Working Paper Series

Differences in the Information Environment Prior to SEOs under Relaxed Disclosure Regulation

Sarah B Clinton Joshua T White Tracie Woidtke

DERA Working Paper 2013-05

NOTE Staff working papers in the DERA Working Paper Series are preliminary materials circulated to stimulate discussion and critical comment References in publications to the DERA Working Paper Series (other than acknowledgement) should be cleared with the author(s) in light of the tentative character of these papers The Securities and Exchange Commission as a matter of policy disclaims responsibility for an y private publication or statement by any of its employees The views expressed herein are those of the author and do not necessarily reflect the views of the Commission or of the authorrsquos colleagues on the staff of the Commission

Differences in the information environment prior to seasoned equity offerings under

relaxed disclosure regulation

Sarah B Clinton

Department of Accounting and Information Management

College of Business Administration

University of Tennessee

620 Stokely Management Center Knoxville TN 37996

clintonutkedu

Joshua T White

US Securities and Exchange Commission Division of Economic and Risk Analysis

Office of Corporate Finance 100 F Street NE Washington DC 20549

whitejosecgov

Tracie Woidtke

Department of Finance

College of Business Administration

University of Tennessee

425 Stokely Management Center Knoxville TN 37996

twoidtkeutkedu

Telephone (865) 974-1718

Fax (865) 974-1716

April 28 2014

We thank an anonymous referee SP Kothari (editor) Anup Agrawal Cindy Alexander Judy

Beckman Bruce Behn James Chyz Shayne Clinton Donial Dastgir Raquel Fox Joan

Heminway Gerard Hoberg Rebekah McCarty Sarah McVay Alex Nekrasov Marlene Plumlee

Andy Puckett Anup Srivastava Amy Starr Karen Ubell Christian Vossler and seminar

participants at the 2010 American Accounting Association Northeast regional meeting 2011

American Accounting Association annual meeting 2012 Financial Accounting and Reporting

Section midyear meeting Southern Methodist University Texas AampM University University of

Alabama University of Arizona Clemson University Miami University University of

Delaware University of Georgia University of Mississippi University of NebraskandashLincoln

University of Utah US Securities and Exchange Commission (SEC) and the University of

Tennessee Corporate Governance Center for insightful comments and suggestions As a matter

of policy the Securities and Exchange Commission disclaims responsibility for any private

publication or statement by any of its employees Although this paper was reviewed by other

staff members within the Commission the views expressed herein are those of the authors and

do not necessarily reflect those of the Commission the Commissioners or other members of the

Staff

Corresponding author

Differences in the information environment prior to seasoned equity offerings under

relaxed disclosure regulation

Sarah B Clinton

Department of Accounting and Information Management

College of Business Administration

University of Tennessee

620 Stokely Management Center Knoxville TN 37996

clintonutkedu

Joshua T White

US Securities and Exchange Commission Division of Economic and Risk Analysis

Office of Corporate Finance 100 F Street NE Washington DC 20549

whitejosecgov

Tracie Woidtke

Department of Finance

College of Business Administration

University of Tennessee

425 Stokely Management Center Knoxville TN 37996

twoidtkeutkedu

Telephone (865) 974-1718

Fax (865) 974-1716

April 28 2014

We thank an anonymous referee SP Kothari (editor) Anup Agrawal Cindy Alexander Judy

Beckman Bruce Behn James Chyz Shayne Clinton Donial Dastgir Raquel Fox Joan

Heminway Gerard Hoberg Rebekah McCarty Sarah McVay Alex Nekrasov Marlene Plumlee

Andy Puckett Anup Srivastava Amy Starr Karen Ubell Christian Vossler and seminar

participants at the 2010 American Accounting Association Northeast regional meeting 2011

American Accounting Association annual meeting 2012 Financial Accounting and Reporting

Section midyear meeting Southern Methodist University Texas AampM University University of

Alabama University of Arizona Clemson University Miami University University of

Delaware University of Georgia University of Mississippi University of NebraskandashLincoln

University of Utah US Securities and Exchange Commission (SEC) and the University of

Tennessee Corporate Governance Center for insightful comments and suggestions As a matter

of policy the Securities and Exchange Commission disclaims responsibility for any private

publication or statement by any of its employees Although this paper was reviewed by other

staff members within the Commission the views expressed herein are those of the authors and

do not necessarily reflect those of the Commission the Commissioners or other members of the

Staff

Corresponding author

Differences in the information environment prior to seasoned equity offerings under

relaxed disclosure regulation

ABSTRACT

The Securities Offering Reform (SOR) promulgated by the US Securities and Exchange

Commission (SEC) in 2005 represents a major change in the seasoned equity offering (SEO)

process SOR eases disclosure restrictions and reduces uncertainty regarding disclosures allowed

prior to an SEO The SEC argued that SOR would result in an improved information

environment and benefit capital formation efficiency In contrast critics claimed that SOR would

allow firms to hype their stock before an SEO to the detriment of the information environment It

is also possible that SOR is not effective in generating greater disclosure and affecting the

information environment during equity capital formation because of ongoing concerns such as

the litigious climate in the US This paper is the first to examine differences in disclosure and

the information environment at the time of seasoned equity capital formation under SOR

Overall the results suggest that relaxed disclosure restrictions under SOR are associated with

greater disclosure immediately preceding the SEO issue date and greater disclosure is related to

a richer information environment with capital formation benefits During the month before the

SEO issue date under SOR we find more frequent disclosure of management earnings forecasts

and Form 8-K filings and management earnings forecasts are more accurate In addition we find

disclosure within a week before the SEO issue date under SOR is associated with greater

absolute market-adjusted returns (ie information magnitude) and more positive stock returns

with no reversal afterward (ie capital formation benefits through informative disclosure)

Keywords Securities Offering Reform Securities regulation Seasoned equity offerings

Disclosure Information environment Shelf registration

JEL classification M41 G38 K22

Differences in the information environment prior to seasoned equity offerings under

relaxed disclosure regulation

1 Introduction

Itrsquos a basic precept of American securities law that shareholders should be given the information

they need to evaluate their companies

- Robert Jackson professor of law Columbia University (New York Times April 24 2013)

The information environment plays a critical role in capital market efficiency However

the regulatory framework and the litigious climate within the US often result in firms disclosing

only information that is required (Cadwalader Wickersham amp Taft LLP 2005 Rogers and Van

Buskirk 2009) Until recently disclosure prior to seasoned equity offerings (SEOs) had been

limited by the US Securities and Exchange Commission (SEC) even though investor demand

for information was high Section 5(c) of the Securities Act restricted firm disclosure before an

SEO to ban any attempts to ldquocondition the marketrdquo The SEC promulgated the Securities

Offering Reform (SOR) in 2005 to modernize the SEO process and eliminate restrictions that

were no longer viewed as necessary to protect investors yet prevented the dissemination of

accurate information1

This paper investigates whether SOR is associated with differences in

disclosure and a richer information environment during the equity capital formation period (ie

the period preceding an SEO issue when investors assess the SEO price and decide whether to

invest)

SOR is a unique shift in the regulatory environment for SEOs In addition to removing

restrictions on factual and forward-looking disclosures prior to SEOs SOR provides a safe

harbor for disclosures and allows a novel disclosure venue the free writing prospectus (FWP)

which is any written communication that offers to sell securities that are or will be subject to a

registration statement and can include factual and forward-looking information Critics argued

1 See the Securities Offering Reform Final Rule at httpwwwsecgovrulesfinal33-8591frpdf

1

Differences in the information environment prior to seasoned equity offerings under

relaxed disclosure regulation

1 Introduction

Itrsquos a basic precept of American securities law that shareholders should be given the information

they need to evaluate their companies

- Robert Jackson professor of law Columbia University (New York Times April 24 2013)

The information environment plays a critical role in capital market efficiency However

the regulatory framework and the litigious climate within the US often result in firms disclosing

only information that is required (Cadwalader Wickersham amp Taft LLP 2005 Rogers and Van

Buskirk 2009) Until recently disclosure prior to seasoned equity offerings (SEOs) had been

limited by the US Securities and Exchange Commission (SEC) even though investor demand

for information was high Section 5(c) of the Securities Act restricted firm disclosure before an

SEO to ban any attempts to ldquocondition the marketrdquo The SEC promulgated the Securities

Offering Reform (SOR) in 2005 to modernize the SEO process and eliminate restrictions that

were no longer viewed as necessary to protect investors yet prevented the dissemination of

accurate information1

This paper investigates whether SOR is associated with differences in

disclosure and a richer information environment during the equity capital formation period (ie

the period preceding an SEO issue when investors assess the SEO price and decide whether to

invest)

SOR is a unique shift in the regulatory environment for SEOs In addition to removing

restrictions on factual and forward-looking disclosures prior to SEOs SOR provides a safe

harbor for disclosures and allows a novel disclosure venue the free writing prospectus (FWP)

which is any written communication that offers to sell securities that are or will be subject to a

registration statement and can include factual and forward-looking information Critics argued

1 See the Securities Offering Reform Final Rule at httpwwwsecgovrulesfinal33-8591frpdf

1

that eased restrictions under SOR would result in firms providing opportunistic disclosure to

ldquohyperdquo expectations before an SEO (eg Morrissey 2007) However the SEC argued that the

primary benefit of SOR was to provide greater and timelier information flow when investors

decide whether to invest in an SEO (SOR Final Rule p 44792) Despite the increased flexibility

under SOR greater disclosure during equity capital formation may not occur for several reasons

Firms may be concerned that greater disclosure could reveal proprietary information (Ali et al

2014) or establish a disclosure precedent that is difficult to maintain or deviates from a long-

standing policy (Graham et al 2005) Firms may also continue the pre-SOR policy of limiting

disclosure during the capital formation period due to liability concerns such as shareholder

litigation or SEC sanctions for conditioning the market (Cadwalader Wickersham amp Taft LLP

2005) Thus it is not clear that SOR automatically leads to greater disclosure or an improved

information environment prior to SEOs

We analyze a sample of 360 SEOs issued by well-known seasoned issuers (WKSIs) a

new class of issuers established by SOR from 2002 to 2009 To provide the first evidence of

whether disclosure is greater and the information environment is richer under SOR during the

capital formation period we conduct tests using multiple forms of factual and forward-looking

disclosure Specifically we examine management earnings forecasts (MEFs) FWPs 8-K filings

and earnings announcements To capture the capital formation period we focus on the month

(ie 30 days) prior to the SEO issue date The role of disclosure before the issue date is

important as a result of changes in the equity offering process including a shift to shelf

registrations and an increase in the distance between the SEO filing and issue dates under SOR

(Table 1) In addition to our overall analysis we separately examine firms with a policy of

providing MEFs as well as other disclosures (forecasting sample) and firms with a policy of

2

providing only disclosures other than MEFs (non-forecasting sample) to investigate whether

firms behave differently under relaxed disclosure regulation based on prior disclosure practices

We conduct three primary tests First we test whether SOR is associated with more

factual and forward-looking disclosure Second we test whether MEF disclosure is indicative of

a richer information environment under SOR for the forecasting sample We examine MEFs to

see if they are more accurate (ie informative) potentially reducing information asymmetry and

litigation risk or overly optimistic (ie opportunistic) as predicted by critics of SOR Third we

analyze stock prices in two ways to further evaluate the information environment and investigate

capital formation benefits We examine absolute market-adjusted returns to measure the

magnitude of information and signed raw returns to investigate whether disclosure is

opportunistic or informative

Overall the results suggest that relaxed disclosure regulation under SOR is associated

with greater disclosure prior to SEOs which is related to a richer information environment with

capital formation benefits Contrary to criticisms of SOR we do not find evidence that these

disclosures are opportunistic in nature The use of FWPs in the full sample and a greater

proportion of annual MEFs in the forecasting sample under SOR result in a 51 greater

proportion of SEOs preceded by guidance disclosure during the capital formation period The

overall frequency of disclosure is 25 greater and the amount of information provided in the 8-

K immediately before the issue date more than doubles in size Furthermore the number of days

between the last disclosure before an SEO and the issue date is lower under SOR Also MEFs

immediately preceding the SEO issue date are 68 more accurate under SOR (ie have reduced

error and bias) and tend to include more downward than upward revisions In addition the

abnormal magnitude of information reflected in absolute market-adjusted returns is almost twice

3

as large after SOR and the difference is driven by SEOs with disclosure within one week before

the issue date Furthermore stock returns during the capital formation period are greater after

SOR for SEOs with disclosure within one week prior to the issue date with no reversal after the

issue date consistent with enhanced capital formation through informative not opportunistic

disclosure These results support the SECrsquos argument that SOR would result in an improved

information environment and benefit capital formation efficiency

Our paper makes several contributions to the literature First we contribute to the

literature on disclosure and the information environment Specifically our paper is the first to

our knowledge to examine different forms of disclosure including the FWP during the equity

capital formation period Second we contribute to the SEO literature by highlighting changes in

the SEO environment that have important implications for future research For example we

show that the capital formation period is different from the pre-SEO period defined in earlier

studies for the majority of SEOs during our more recent sample period Although earlier studies

frequently designate the filing date as the SEO announcement date (eg Asquith and Mullins

1986 Masulis and Korwar 1986) because few SEOs were shelf-registered studies covering

more recent periods and using the filing date as the SEO announcement date may misinterpret

insignificant returns as less negative reactions to SEOs Less information is conveyed at the

filing date as a result of SOR Finally we contribute to research on regulatory reform While

other studies on relaxed regulation tend to find negative outcomes our results indicate that SOR

increases information flow and smoothes capital formation2

Our paper advances the literature on

disclosure SEOs and regulatory reform and should be of interest to the SEC and firms that

access capital on the secondary market

2 Empirical evidence of negative outcomes is linked to relaxed regulation in disclosure (eg Fernandes et al 2010)

and financial markets (eg Beck et al 2010)

4

The remainder of the paper is organized as follows Section 2 provides a discussion of

SOR and presents our empirical questions Section 3 describes the data and Section 4 outlines

our research design Section 5 presents our results Section 6 concludes the paper

2 Background and empirical questions

21 Before the Securities Offering Reform of 2005

Before SOR the SEO process was commonly divided into three periods pre-filing

waiting and post-effective (see Fig 1 for shelf registration timeline) The pre-filing period began

when a firm contemplated an SEO and ended at the filing date when the registration statement

was filed with the SEC The waiting period began at the filing date and ended when the SEC

declared the registration statement effective During the post-effective period a shelf-registered

SEO firm usually filed a preliminary prospectus (pre-issue date) before the shares were sold

(issue date) The post-effective period was typically short to nonexistent for non-shelf (ie

traditional) registrations because the filing date which was also the pre-issue date and issue date

were very close (Bethel and Krigman 2008)

[Insert Fig 1 near here]

Prior to the reform certain communication was prohibited in the pre-filing and waiting

periods and was limited in the post-effective period The pre-filing and waiting periods were

covered by disclosure restrictions banning any attempt to condition the market Any additional

disclosures in connection with the SEO including forward-looking statements were expressly

forbidden with the exception of a prospectus During the post-effective period factual and

forward-looking disclosure was effectively limited through cumbersome prospectus amendment

requirements and liability concerns (Latham amp Watkins LLP 2005) For example forecasts had

5

to be filed as part of the prospectus Many issuers did not want to include a forecast in a

prospectus because it could be viewed as conditioning the market3

Any disclosure that could be

viewed as an offer (eg an MEF) was limited to statutory prospectus disclosure Furthermore

once the final prospectus was prepared issuers were responsible for ensuring that an investor had

received the final prospectus and the SEC reserved the right to review the final prospectus

which could potentially delay the SEO

22 After the Securities Offering Reform of 2005

SOR became effective December 1 2005 Although SOR affects all issuers it provides

the most significant changes for a newly established class of issuers WKSIs which have a

market capitalization of at least $700 million4

The rationale for WKSIs receiving the most

flexibility is that these firms are so widely followed that investors need less regulatory

protection WKSIs accounted for 95 of the US market capitalization during 2002 to 2009

Even though WKSIs include the largest SEO firms 82 of our sample is composed of mid-cap

(447) and small-cap (375) firms (untabulated)5

Instead of restricting disclosure the SEC

argues that disclosure should be encouraged for these firms (SOR Final Rule p 44731)

SOR relaxes disclosure restrictions for WKSIs in several ways First it removes the

waiting period and changes the pre-filing and post-effective periods by dropping communication

restrictions throughout the entire SEO process (see Fig 1) WKSIs may report in any period

3 Some attorneys post-SOR continued to advise against providing any communications at any time before the SEO

that could be viewed as preparing the market for an offering or that had not undergone due diligence in a prospectus

(Cadwalader Wickersham amp Taft LLP 2005) 4

Alternatively firms with at least $1 billion of nonconvertible securities issuances over the past three years qualify

as WKSIs 5

These figures use the following thresholds for market capitalization $07 billion to $14 billion are small -cap $15

billion to $44 billion are mid-cap $45 billion and above are large-cap

6

factual and forward-looking information that the firm regularly releases6

Second SOR provides

multiple safe harbors for regularly released factual business and forward-looking information7

Third SOR allows WKSIs to freely communicate (ie free write) during any period using an

FWP FWPs include communications in written form and communications via television radio

broadcasts and certain electronic road show slides or handouts (Pena 2007)8

Fourth for

WKSIs shelf registrations are automatically effective without SEC review and the filing period

for shelf registrations is extended from two to three years WKSIs can also update or renew the

shelf registration without SEC review and are no longer responsible for ensuring that investors

receive the final prospectus because filing with the SEC is now viewed as ldquoaccess equals

deliveryrdquo (SOR Final Rule pp 44784-44785) These changes increase a WKSIrsquos ability to

quickly access capital markets when necessary Finally SOR allows firms to omit information

(eg proposed offering amounts for each type of security) from the shelf registration which

effectively shifts the SEO information event from the filing date to the pre-issue date9

As Fig 1

indicates the shelf registration Form S-3ASR allows WKSIs to register to offer securities at any

time without stating detailed information about the SEO or paying registration fees which would

provide some indication of the offering amount10

6 In the SOR Final Rule the SEC clarifies that regularly released forecasts include scheduled forecasts as well as

unscheduled or episodic forecasts The firm could satisfy the regularly released condition as long as the company

has released one prior forecast (SOR Final Rule p 44737) However Cadwalader Wickersham amp Taft LLP a large

international law firm included the following statement in its August 4 2005 memo explaining SOR ldquoThe safe harbors do not establish any minimum time period to satisfy the lsquoregularly releasedrsquo element of the Final Rules While the SEC has stated that one prior release or dissemination could establish a track record if such

communication is the industry norm we believe that commencing a practice for the purpose of meeting this

requirement of the safe harbor will have significant risk until there is more clarification on this pointrdquo (Cadwalader Wickersham amp Taft LLP p 6) 7 Under SORrsquos safe harbor WKSIs are not liable for any forward-looking statement as long as it is accompanied by

certain cautionary language and the WKSI did not have knowledge that the statement was false 8

FWP filings are not required for traditional live road shows in which the SEO firm and underwriters meet potential investors 9

Autore et al (2008) report that 85 of firms filing a shelf registration never issue equity 10

For example firms are not required to state the size of the offering or indicate the specific security that will be issued at the filing date Before SOR the shelf registration included the amount that firms expected to issue over the

7

providing only disclosures other than MEFs (non-forecasting sample) to investigate whether

firms behave differently under relaxed disclosure regulation based on prior disclosure practices

We conduct three primary tests First we test whether SOR is associated with more

factual and forward-looking disclosure Second we test whether MEF disclosure is indicative of

a richer information environment under SOR for the forecasting sample We examine MEFs to

see if they are more accurate (ie informative) potentially reducing information asymmetry and

litigation risk or overly optimistic (ie opportunistic) as predicted by critics of SOR Third we

analyze stock prices in two ways to further evaluate the information environment and investigate

capital formation benefits We examine absolute market-adjusted returns to measure the

magnitude of information and signed raw returns to investigate whether disclosure is

opportunistic or informative

Overall the results suggest that relaxed disclosure regulation under SOR is associated

with greater disclosure prior to SEOs which is related to a richer information environment with

capital formation benefits Contrary to criticisms of SOR we do not find evidence that these

disclosures are opportunistic in nature The use of FWPs in the full sample and a greater

proportion of annual MEFs in the forecasting sample under SOR result in a 51 greater

proportion of SEOs preceded by guidance disclosure during the capital formation period The

overall frequency of disclosure is 25 greater and the amount of information provided in the 8-

K immediately before the issue date more than doubles in size Furthermore the number of days

between the last disclosure before an SEO and the issue date is lower under SOR Also MEFs

immediately preceding the SEO issue date are 68 more accurate under SOR (ie have reduced

error and bias) and tend to include more downward than upward revisions In addition the

abnormal magnitude of information reflected in absolute market-adjusted returns is almost twice

3

as large after SOR and the difference is driven by SEOs with disclosure within one week before

the issue date Furthermore stock returns during the capital formation period are greater after

SOR for SEOs with disclosure within one week prior to the issue date with no reversal after the

issue date consistent with enhanced capital formation through informative not opportunistic

disclosure These results support the SECrsquos argument that SOR would result in an improved

information environment and benefit capital formation efficiency

Our paper makes several contributions to the literature First we contribute to the

literature on disclosure and the information environment Specifically our paper is the first to

our knowledge to examine different forms of disclosure including the FWP during the equity

capital formation period Second we contribute to the SEO literature by highlighting changes in

the SEO environment that have important implications for future research For example we

show that the capital formation period is different from the pre-SEO period defined in earlier

studies for the majority of SEOs during our more recent sample period Although earlier studies

frequently designate the filing date as the SEO announcement date (eg Asquith and Mullins

1986 Masulis and Korwar 1986) because few SEOs were shelf-registered studies covering

more recent periods and using the filing date as the SEO announcement date may misinterpret

insignificant returns as less negative reactions to SEOs Less information is conveyed at the

filing date as a result of SOR Finally we contribute to research on regulatory reform While

other studies on relaxed regulation tend to find negative outcomes our results indicate that SOR

increases information flow and smoothes capital formation2

Our paper advances the literature on

disclosure SEOs and regulatory reform and should be of interest to the SEC and firms that

access capital on the secondary market

2 Empirical evidence of negative outcomes is linked to relaxed regulation in disclosure (eg Fernandes et al 2010)

and financial markets (eg Beck et al 2010)

4

The remainder of the paper is organized as follows Section 2 provides a discussion of

SOR and presents our empirical questions Section 3 describes the data and Section 4 outlines

our research design Section 5 presents our results Section 6 concludes the paper

2 Background and empirical questions

21 Before the Securities Offering Reform of 2005

Before SOR the SEO process was commonly divided into three periods pre-filing

waiting and post-effective (see Fig 1 for shelf registration timeline) The pre-filing period began

when a firm contemplated an SEO and ended at the filing date when the registration statement

was filed with the SEC The waiting period began at the filing date and ended when the SEC

declared the registration statement effective During the post-effective period a shelf-registered

SEO firm usually filed a preliminary prospectus (pre-issue date) before the shares were sold

(issue date) The post-effective period was typically short to nonexistent for non-shelf (ie

traditional) registrations because the filing date which was also the pre-issue date and issue date

were very close (Bethel and Krigman 2008)

[Insert Fig 1 near here]

Prior to the reform certain communication was prohibited in the pre-filing and waiting

periods and was limited in the post-effective period The pre-filing and waiting periods were

covered by disclosure restrictions banning any attempt to condition the market Any additional

disclosures in connection with the SEO including forward-looking statements were expressly

forbidden with the exception of a prospectus During the post-effective period factual and

forward-looking disclosure was effectively limited through cumbersome prospectus amendment

requirements and liability concerns (Latham amp Watkins LLP 2005) For example forecasts had

5

to be filed as part of the prospectus Many issuers did not want to include a forecast in a

prospectus because it could be viewed as conditioning the market3

Any disclosure that could be

viewed as an offer (eg an MEF) was limited to statutory prospectus disclosure Furthermore

once the final prospectus was prepared issuers were responsible for ensuring that an investor had

received the final prospectus and the SEC reserved the right to review the final prospectus

which could potentially delay the SEO

22 After the Securities Offering Reform of 2005

SOR became effective December 1 2005 Although SOR affects all issuers it provides

the most significant changes for a newly established class of issuers WKSIs which have a

market capitalization of at least $700 million4

The rationale for WKSIs receiving the most

flexibility is that these firms are so widely followed that investors need less regulatory

protection WKSIs accounted for 95 of the US market capitalization during 2002 to 2009

Even though WKSIs include the largest SEO firms 82 of our sample is composed of mid-cap

(447) and small-cap (375) firms (untabulated)5

Instead of restricting disclosure the SEC

argues that disclosure should be encouraged for these firms (SOR Final Rule p 44731)

SOR relaxes disclosure restrictions for WKSIs in several ways First it removes the

waiting period and changes the pre-filing and post-effective periods by dropping communication

restrictions throughout the entire SEO process (see Fig 1) WKSIs may report in any period

3 Some attorneys post-SOR continued to advise against providing any communications at any time before the SEO

that could be viewed as preparing the market for an offering or that had not undergone due diligence in a prospectus

(Cadwalader Wickersham amp Taft LLP 2005) 4

Alternatively firms with at least $1 billion of nonconvertible securities issuances over the past three years qualify

as WKSIs 5

These figures use the following thresholds for market capitalization $07 billion to $14 billion are small -cap $15

billion to $44 billion are mid-cap $45 billion and above are large-cap

6

factual and forward-looking information that the firm regularly releases6

Second SOR provides

multiple safe harbors for regularly released factual business and forward-looking information7

Third SOR allows WKSIs to freely communicate (ie free write) during any period using an

FWP FWPs include communications in written form and communications via television radio

broadcasts and certain electronic road show slides or handouts (Pena 2007)8

Fourth for

WKSIs shelf registrations are automatically effective without SEC review and the filing period

for shelf registrations is extended from two to three years WKSIs can also update or renew the

shelf registration without SEC review and are no longer responsible for ensuring that investors

receive the final prospectus because filing with the SEC is now viewed as ldquoaccess equals

deliveryrdquo (SOR Final Rule pp 44784-44785) These changes increase a WKSIrsquos ability to

quickly access capital markets when necessary Finally SOR allows firms to omit information

(eg proposed offering amounts for each type of security) from the shelf registration which

effectively shifts the SEO information event from the filing date to the pre-issue date9

As Fig 1

indicates the shelf registration Form S-3ASR allows WKSIs to register to offer securities at any

time without stating detailed information about the SEO or paying registration fees which would

provide some indication of the offering amount10

6 In the SOR Final Rule the SEC clarifies that regularly released forecasts include scheduled forecasts as well as

unscheduled or episodic forecasts The firm could satisfy the regularly released condition as long as the company

has released one prior forecast (SOR Final Rule p 44737) However Cadwalader Wickersham amp Taft LLP a large

international law firm included the following statement in its August 4 2005 memo explaining SOR ldquoThe safe harbors do not establish any minimum time period to satisfy the lsquoregularly releasedrsquo element of the Final Rules While the SEC has stated that one prior release or dissemination could establish a track record if such

communication is the industry norm we believe that commencing a practice for the purpose of meeting this

requirement of the safe harbor will have significant risk until there is more clarification on this pointrdquo (Cadwalader Wickersham amp Taft LLP p 6) 7 Under SORrsquos safe harbor WKSIs are not liable for any forward-looking statement as long as it is accompanied by

certain cautionary language and the WKSI did not have knowledge that the statement was false 8

FWP filings are not required for traditional live road shows in which the SEO firm and underwriters meet potential investors 9

Autore et al (2008) report that 85 of firms filing a shelf registration never issue equity 10

For example firms are not required to state the size of the offering or indicate the specific security that will be issued at the filing date Before SOR the shelf registration included the amount that firms expected to issue over the

7

as large after SOR and the difference is driven by SEOs with disclosure within one week before

the issue date Furthermore stock returns during the capital formation period are greater after

SOR for SEOs with disclosure within one week prior to the issue date with no reversal after the

issue date consistent with enhanced capital formation through informative not opportunistic

disclosure These results support the SECrsquos argument that SOR would result in an improved

information environment and benefit capital formation efficiency

Our paper makes several contributions to the literature First we contribute to the

literature on disclosure and the information environment Specifically our paper is the first to

our knowledge to examine different forms of disclosure including the FWP during the equity

capital formation period Second we contribute to the SEO literature by highlighting changes in

the SEO environment that have important implications for future research For example we

show that the capital formation period is different from the pre-SEO period defined in earlier

studies for the majority of SEOs during our more recent sample period Although earlier studies

frequently designate the filing date as the SEO announcement date (eg Asquith and Mullins

1986 Masulis and Korwar 1986) because few SEOs were shelf-registered studies covering

more recent periods and using the filing date as the SEO announcement date may misinterpret

insignificant returns as less negative reactions to SEOs Less information is conveyed at the

filing date as a result of SOR Finally we contribute to research on regulatory reform While

other studies on relaxed regulation tend to find negative outcomes our results indicate that SOR

increases information flow and smoothes capital formation2

Our paper advances the literature on

disclosure SEOs and regulatory reform and should be of interest to the SEC and firms that

access capital on the secondary market

2 Empirical evidence of negative outcomes is linked to relaxed regulation in disclosure (eg Fernandes et al 2010)

and financial markets (eg Beck et al 2010)

4

The remainder of the paper is organized as follows Section 2 provides a discussion of

SOR and presents our empirical questions Section 3 describes the data and Section 4 outlines

our research design Section 5 presents our results Section 6 concludes the paper

2 Background and empirical questions

21 Before the Securities Offering Reform of 2005

Before SOR the SEO process was commonly divided into three periods pre-filing

waiting and post-effective (see Fig 1 for shelf registration timeline) The pre-filing period began

when a firm contemplated an SEO and ended at the filing date when the registration statement

was filed with the SEC The waiting period began at the filing date and ended when the SEC

declared the registration statement effective During the post-effective period a shelf-registered

SEO firm usually filed a preliminary prospectus (pre-issue date) before the shares were sold

(issue date) The post-effective period was typically short to nonexistent for non-shelf (ie

traditional) registrations because the filing date which was also the pre-issue date and issue date

were very close (Bethel and Krigman 2008)

[Insert Fig 1 near here]

Prior to the reform certain communication was prohibited in the pre-filing and waiting

periods and was limited in the post-effective period The pre-filing and waiting periods were

covered by disclosure restrictions banning any attempt to condition the market Any additional

disclosures in connection with the SEO including forward-looking statements were expressly

forbidden with the exception of a prospectus During the post-effective period factual and

forward-looking disclosure was effectively limited through cumbersome prospectus amendment

requirements and liability concerns (Latham amp Watkins LLP 2005) For example forecasts had

5

to be filed as part of the prospectus Many issuers did not want to include a forecast in a

prospectus because it could be viewed as conditioning the market3

Any disclosure that could be

viewed as an offer (eg an MEF) was limited to statutory prospectus disclosure Furthermore

once the final prospectus was prepared issuers were responsible for ensuring that an investor had

received the final prospectus and the SEC reserved the right to review the final prospectus

which could potentially delay the SEO

22 After the Securities Offering Reform of 2005

SOR became effective December 1 2005 Although SOR affects all issuers it provides

the most significant changes for a newly established class of issuers WKSIs which have a

market capitalization of at least $700 million4

The rationale for WKSIs receiving the most

flexibility is that these firms are so widely followed that investors need less regulatory

protection WKSIs accounted for 95 of the US market capitalization during 2002 to 2009

Even though WKSIs include the largest SEO firms 82 of our sample is composed of mid-cap

(447) and small-cap (375) firms (untabulated)5

Instead of restricting disclosure the SEC

argues that disclosure should be encouraged for these firms (SOR Final Rule p 44731)

SOR relaxes disclosure restrictions for WKSIs in several ways First it removes the

waiting period and changes the pre-filing and post-effective periods by dropping communication

restrictions throughout the entire SEO process (see Fig 1) WKSIs may report in any period

3 Some attorneys post-SOR continued to advise against providing any communications at any time before the SEO

that could be viewed as preparing the market for an offering or that had not undergone due diligence in a prospectus

(Cadwalader Wickersham amp Taft LLP 2005) 4

Alternatively firms with at least $1 billion of nonconvertible securities issuances over the past three years qualify

as WKSIs 5

These figures use the following thresholds for market capitalization $07 billion to $14 billion are small -cap $15

billion to $44 billion are mid-cap $45 billion and above are large-cap

6

factual and forward-looking information that the firm regularly releases6

Second SOR provides

multiple safe harbors for regularly released factual business and forward-looking information7

Third SOR allows WKSIs to freely communicate (ie free write) during any period using an

FWP FWPs include communications in written form and communications via television radio

broadcasts and certain electronic road show slides or handouts (Pena 2007)8

Fourth for

WKSIs shelf registrations are automatically effective without SEC review and the filing period

for shelf registrations is extended from two to three years WKSIs can also update or renew the

shelf registration without SEC review and are no longer responsible for ensuring that investors

receive the final prospectus because filing with the SEC is now viewed as ldquoaccess equals

deliveryrdquo (SOR Final Rule pp 44784-44785) These changes increase a WKSIrsquos ability to

quickly access capital markets when necessary Finally SOR allows firms to omit information

(eg proposed offering amounts for each type of security) from the shelf registration which

effectively shifts the SEO information event from the filing date to the pre-issue date9

As Fig 1

indicates the shelf registration Form S-3ASR allows WKSIs to register to offer securities at any

time without stating detailed information about the SEO or paying registration fees which would

provide some indication of the offering amount10

6 In the SOR Final Rule the SEC clarifies that regularly released forecasts include scheduled forecasts as well as

unscheduled or episodic forecasts The firm could satisfy the regularly released condition as long as the company

has released one prior forecast (SOR Final Rule p 44737) However Cadwalader Wickersham amp Taft LLP a large

international law firm included the following statement in its August 4 2005 memo explaining SOR ldquoThe safe harbors do not establish any minimum time period to satisfy the lsquoregularly releasedrsquo element of the Final Rules While the SEC has stated that one prior release or dissemination could establish a track record if such

communication is the industry norm we believe that commencing a practice for the purpose of meeting this

requirement of the safe harbor will have significant risk until there is more clarification on this pointrdquo (Cadwalader Wickersham amp Taft LLP p 6) 7 Under SORrsquos safe harbor WKSIs are not liable for any forward-looking statement as long as it is accompanied by

certain cautionary language and the WKSI did not have knowledge that the statement was false 8

FWP filings are not required for traditional live road shows in which the SEO firm and underwriters meet potential investors 9

Autore et al (2008) report that 85 of firms filing a shelf registration never issue equity 10

For example firms are not required to state the size of the offering or indicate the specific security that will be issued at the filing date Before SOR the shelf registration included the amount that firms expected to issue over the

7

The remainder of the paper is organized as follows Section 2 provides a discussion of

SOR and presents our empirical questions Section 3 describes the data and Section 4 outlines

our research design Section 5 presents our results Section 6 concludes the paper

2 Background and empirical questions

21 Before the Securities Offering Reform of 2005

Before SOR the SEO process was commonly divided into three periods pre-filing

waiting and post-effective (see Fig 1 for shelf registration timeline) The pre-filing period began

when a firm contemplated an SEO and ended at the filing date when the registration statement

was filed with the SEC The waiting period began at the filing date and ended when the SEC

declared the registration statement effective During the post-effective period a shelf-registered

SEO firm usually filed a preliminary prospectus (pre-issue date) before the shares were sold

(issue date) The post-effective period was typically short to nonexistent for non-shelf (ie

traditional) registrations because the filing date which was also the pre-issue date and issue date

were very close (Bethel and Krigman 2008)

[Insert Fig 1 near here]

Prior to the reform certain communication was prohibited in the pre-filing and waiting

periods and was limited in the post-effective period The pre-filing and waiting periods were

covered by disclosure restrictions banning any attempt to condition the market Any additional

disclosures in connection with the SEO including forward-looking statements were expressly

forbidden with the exception of a prospectus During the post-effective period factual and

forward-looking disclosure was effectively limited through cumbersome prospectus amendment

requirements and liability concerns (Latham amp Watkins LLP 2005) For example forecasts had

5

to be filed as part of the prospectus Many issuers did not want to include a forecast in a

prospectus because it could be viewed as conditioning the market3

Any disclosure that could be

viewed as an offer (eg an MEF) was limited to statutory prospectus disclosure Furthermore

once the final prospectus was prepared issuers were responsible for ensuring that an investor had

received the final prospectus and the SEC reserved the right to review the final prospectus

which could potentially delay the SEO

22 After the Securities Offering Reform of 2005

SOR became effective December 1 2005 Although SOR affects all issuers it provides

the most significant changes for a newly established class of issuers WKSIs which have a

market capitalization of at least $700 million4

The rationale for WKSIs receiving the most

flexibility is that these firms are so widely followed that investors need less regulatory

protection WKSIs accounted for 95 of the US market capitalization during 2002 to 2009

Even though WKSIs include the largest SEO firms 82 of our sample is composed of mid-cap

(447) and small-cap (375) firms (untabulated)5

Instead of restricting disclosure the SEC

argues that disclosure should be encouraged for these firms (SOR Final Rule p 44731)

SOR relaxes disclosure restrictions for WKSIs in several ways First it removes the

waiting period and changes the pre-filing and post-effective periods by dropping communication

restrictions throughout the entire SEO process (see Fig 1) WKSIs may report in any period

3 Some attorneys post-SOR continued to advise against providing any communications at any time before the SEO

that could be viewed as preparing the market for an offering or that had not undergone due diligence in a prospectus

(Cadwalader Wickersham amp Taft LLP 2005) 4

Alternatively firms with at least $1 billion of nonconvertible securities issuances over the past three years qualify

as WKSIs 5

These figures use the following thresholds for market capitalization $07 billion to $14 billion are small -cap $15

billion to $44 billion are mid-cap $45 billion and above are large-cap

6

factual and forward-looking information that the firm regularly releases6

Second SOR provides

multiple safe harbors for regularly released factual business and forward-looking information7

Third SOR allows WKSIs to freely communicate (ie free write) during any period using an

FWP FWPs include communications in written form and communications via television radio

broadcasts and certain electronic road show slides or handouts (Pena 2007)8

Fourth for

WKSIs shelf registrations are automatically effective without SEC review and the filing period

for shelf registrations is extended from two to three years WKSIs can also update or renew the

shelf registration without SEC review and are no longer responsible for ensuring that investors

receive the final prospectus because filing with the SEC is now viewed as ldquoaccess equals

deliveryrdquo (SOR Final Rule pp 44784-44785) These changes increase a WKSIrsquos ability to

quickly access capital markets when necessary Finally SOR allows firms to omit information

(eg proposed offering amounts for each type of security) from the shelf registration which

effectively shifts the SEO information event from the filing date to the pre-issue date9

As Fig 1

indicates the shelf registration Form S-3ASR allows WKSIs to register to offer securities at any

time without stating detailed information about the SEO or paying registration fees which would

provide some indication of the offering amount10

6 In the SOR Final Rule the SEC clarifies that regularly released forecasts include scheduled forecasts as well as

unscheduled or episodic forecasts The firm could satisfy the regularly released condition as long as the company

has released one prior forecast (SOR Final Rule p 44737) However Cadwalader Wickersham amp Taft LLP a large

international law firm included the following statement in its August 4 2005 memo explaining SOR ldquoThe safe harbors do not establish any minimum time period to satisfy the lsquoregularly releasedrsquo element of the Final Rules While the SEC has stated that one prior release or dissemination could establish a track record if such

communication is the industry norm we believe that commencing a practice for the purpose of meeting this

requirement of the safe harbor will have significant risk until there is more clarification on this pointrdquo (Cadwalader Wickersham amp Taft LLP p 6) 7 Under SORrsquos safe harbor WKSIs are not liable for any forward-looking statement as long as it is accompanied by

certain cautionary language and the WKSI did not have knowledge that the statement was false 8

FWP filings are not required for traditional live road shows in which the SEO firm and underwriters meet potential investors 9

Autore et al (2008) report that 85 of firms filing a shelf registration never issue equity 10

For example firms are not required to state the size of the offering or indicate the specific security that will be issued at the filing date Before SOR the shelf registration included the amount that firms expected to issue over the

7

to be filed as part of the prospectus Many issuers did not want to include a forecast in a

prospectus because it could be viewed as conditioning the market3

Any disclosure that could be

viewed as an offer (eg an MEF) was limited to statutory prospectus disclosure Furthermore

once the final prospectus was prepared issuers were responsible for ensuring that an investor had

received the final prospectus and the SEC reserved the right to review the final prospectus

which could potentially delay the SEO

22 After the Securities Offering Reform of 2005

SOR became effective December 1 2005 Although SOR affects all issuers it provides

the most significant changes for a newly established class of issuers WKSIs which have a

market capitalization of at least $700 million4

The rationale for WKSIs receiving the most

flexibility is that these firms are so widely followed that investors need less regulatory

protection WKSIs accounted for 95 of the US market capitalization during 2002 to 2009

Even though WKSIs include the largest SEO firms 82 of our sample is composed of mid-cap

(447) and small-cap (375) firms (untabulated)5

Instead of restricting disclosure the SEC

argues that disclosure should be encouraged for these firms (SOR Final Rule p 44731)

SOR relaxes disclosure restrictions for WKSIs in several ways First it removes the

waiting period and changes the pre-filing and post-effective periods by dropping communication

restrictions throughout the entire SEO process (see Fig 1) WKSIs may report in any period

3 Some attorneys post-SOR continued to advise against providing any communications at any time before the SEO

that could be viewed as preparing the market for an offering or that had not undergone due diligence in a prospectus

(Cadwalader Wickersham amp Taft LLP 2005) 4

Alternatively firms with at least $1 billion of nonconvertible securities issuances over the past three years qualify

as WKSIs 5

These figures use the following thresholds for market capitalization $07 billion to $14 billion are small -cap $15

billion to $44 billion are mid-cap $45 billion and above are large-cap

6

factual and forward-looking information that the firm regularly releases6

Second SOR provides

multiple safe harbors for regularly released factual business and forward-looking information7

Third SOR allows WKSIs to freely communicate (ie free write) during any period using an

FWP FWPs include communications in written form and communications via television radio

broadcasts and certain electronic road show slides or handouts (Pena 2007)8

Fourth for

WKSIs shelf registrations are automatically effective without SEC review and the filing period

for shelf registrations is extended from two to three years WKSIs can also update or renew the

shelf registration without SEC review and are no longer responsible for ensuring that investors

receive the final prospectus because filing with the SEC is now viewed as ldquoaccess equals

deliveryrdquo (SOR Final Rule pp 44784-44785) These changes increase a WKSIrsquos ability to

quickly access capital markets when necessary Finally SOR allows firms to omit information

(eg proposed offering amounts for each type of security) from the shelf registration which

effectively shifts the SEO information event from the filing date to the pre-issue date9

As Fig 1

indicates the shelf registration Form S-3ASR allows WKSIs to register to offer securities at any

time without stating detailed information about the SEO or paying registration fees which would

provide some indication of the offering amount10

6 In the SOR Final Rule the SEC clarifies that regularly released forecasts include scheduled forecasts as well as

unscheduled or episodic forecasts The firm could satisfy the regularly released condition as long as the company

has released one prior forecast (SOR Final Rule p 44737) However Cadwalader Wickersham amp Taft LLP a large

international law firm included the following statement in its August 4 2005 memo explaining SOR ldquoThe safe harbors do not establish any minimum time period to satisfy the lsquoregularly releasedrsquo element of the Final Rules While the SEC has stated that one prior release or dissemination could establish a track record if such

communication is the industry norm we believe that commencing a practice for the purpose of meeting this

requirement of the safe harbor will have significant risk until there is more clarification on this pointrdquo (Cadwalader Wickersham amp Taft LLP p 6) 7 Under SORrsquos safe harbor WKSIs are not liable for any forward-looking statement as long as it is accompanied by

certain cautionary language and the WKSI did not have knowledge that the statement was false 8

FWP filings are not required for traditional live road shows in which the SEO firm and underwriters meet potential investors 9

Autore et al (2008) report that 85 of firms filing a shelf registration never issue equity 10

For example firms are not required to state the size of the offering or indicate the specific security that will be issued at the filing date Before SOR the shelf registration included the amount that firms expected to issue over the

7

factual and forward-looking information that the firm regularly releases6

Second SOR provides

multiple safe harbors for regularly released factual business and forward-looking information7

Third SOR allows WKSIs to freely communicate (ie free write) during any period using an

FWP FWPs include communications in written form and communications via television radio

broadcasts and certain electronic road show slides or handouts (Pena 2007)8

Fourth for

WKSIs shelf registrations are automatically effective without SEC review and the filing period

for shelf registrations is extended from two to three years WKSIs can also update or renew the

shelf registration without SEC review and are no longer responsible for ensuring that investors

receive the final prospectus because filing with the SEC is now viewed as ldquoaccess equals

deliveryrdquo (SOR Final Rule pp 44784-44785) These changes increase a WKSIrsquos ability to

quickly access capital markets when necessary Finally SOR allows firms to omit information

(eg proposed offering amounts for each type of security) from the shelf registration which

effectively shifts the SEO information event from the filing date to the pre-issue date9

As Fig 1

indicates the shelf registration Form S-3ASR allows WKSIs to register to offer securities at any

time without stating detailed information about the SEO or paying registration fees which would

provide some indication of the offering amount10

6 In the SOR Final Rule the SEC clarifies that regularly released forecasts include scheduled forecasts as well as

unscheduled or episodic forecasts The firm could satisfy the regularly released condition as long as the company

has released one prior forecast (SOR Final Rule p 44737) However Cadwalader Wickersham amp Taft LLP a large

international law firm included the following statement in its August 4 2005 memo explaining SOR ldquoThe safe harbors do not establish any minimum time period to satisfy the lsquoregularly releasedrsquo element of the Final Rules While the SEC has stated that one prior release or dissemination could establish a track record if such

communication is the industry norm we believe that commencing a practice for the purpose of meeting this

requirement of the safe harbor will have significant risk until there is more clarification on this pointrdquo (Cadwalader Wickersham amp Taft LLP p 6) 7 Under SORrsquos safe harbor WKSIs are not liable for any forward-looking statement as long as it is accompanied by

certain cautionary language and the WKSI did not have knowledge that the statement was false 8

FWP filings are not required for traditional live road shows in which the SEO firm and underwriters meet potential investors 9

Autore et al (2008) report that 85 of firms filing a shelf registration never issue equity 10

For example firms are not required to state the size of the offering or indicate the specific security that will be issued at the filing date Before SOR the shelf registration included the amount that firms expected to issue over the

7

23 Empirical tests

We investigate whether SOR is associated with greater disclosure and a richer

information environment during the capital formation period through three primary tests First

we investigate whether SOR is associated with more factual and forward-looking disclosure by

comparing different measures of disclosure activity (disclosure occurrence frequency 8-K size

and timing) during the capital formation period before and after SOR was promulgated Prior

research provides several reasons for occurrence frequency and size of disclosure to be greater

during this period deter certain types of litigation (eg Field et al 2005) reduce underpricing

(eg Li and Zhuang 2012) lower equity costs (eg Botosan 1997) lower information

asymmetry (eg Diamond and Verrecchia 1991) continue a prior commitment to disclosure

(eg Graham et al 2005) and condition the market (eg Lang and Lundholm 2000) On the

one hand disclosure should increase with relaxed disclosure restrictions if the benefits of greater

disclosure outweigh the costs under SOR Moreover a firm concerned with omitting material

information may provide greater disclosure (SOR Final Rule pp 44795ndash44796) On the other

hand firms may choose to maintain their existing limited disclosure policy and thus not provide

greater disclosure during the capital formation period under the new rules Firms may be

concerned with inadvertent material misstatements of factual or forward-looking information

leading to litigation Legal firms may also advise firms to disclose no more than what is required

during the capital formation period (Cadwalader Wickersham amp Taft LLP 2005)

We also examine the proximity between disclosure of MEFs FWPs 8-Ks and earnings

announcements and the SEO issue date to investigate whether firms alter the disclosure or SEO

next two years and the registration fees paid on this amount Moreover many pre-SOR firms in our sample list only

one or two types of securities such as common stock and debt securities Many WKSIs in our post-SOR sample list

numerous securities in their shelf registration that have never been issued Thus investors do not know how many

shares of equity a firm may issue until the firm files the preliminary prospectus which (if one is filed) discloses the

SEO size but not price or the final prospectus

8

timing or both to affect the distance (ie number of days) under SOR Although firms have

specific deadlines for reporting items on an 8-K evidence by Segal and Segal (2013) shows that

managers engage in strategic timing Furthermore the results in Korajczyk et al (1991) suggest

that firms especially non-forecasting firms strategically time SEOs just after earnings

announcements because they prefer issuing equity when investors are most informed about firm

quality However firms may be less dependent upon earnings announcements (ie greater

distance) since they can provide additional disclosure prior to an SEO under SOR

Second we examine whether more MEFs within the capital formation period is

indicative of a richer information environment under SOR by investigating the accuracy of

MEFs Firms have dueling incentives to provide accurate MEFs prior to an SEO Firms may be

able to issue stock at a higher price following opportunistic disclosure (Lang and Lundholm

2000) However Li and Zhuang (2012) find that highly accurate MEFs reduce the cost of an

SEO MEF accuracy during the month prior to an SEO may also be subject to heightened

scrutiny and litigation risk despite the new safe harbor provisions Improved accuracy would be

consistent with informative voluntary disclosure during the capital formation period while worse

accuracy would be consistent with opportunistic voluntary disclosure as predicted by SORrsquos

critics (eg Morrissey 2007)

Finally we examine two aspects of the information environment reflected in stock prices

First we examine whether the magnitude of information measured as cumulative absolute

market-adjusted returns is greater during the capital formation period All else equal if

disclosure during the capital formation period is associated with greater information absolute

market-adjusted returns should be larger during the month prior to the SEO under SOR If a

larger information magnitude results from greater disclosure immediately preceding the issue

9

date then the information magnitude should become larger as the issue date approaches and for

firms providing disclosure within one week prior to the SEO Second we examine whether SOR

is associated with capital formation benefits by cumulating signed raw returns during the capital

formation period To the extent that SOR is associated with increased offering flexibility through

enhanced shelf registration and communication mechanisms SOR should be associated with

capital formation benefits reflected in higher cumulative stock returns during the capital

formation period ie a larger increase in stock price prior to the issue date We examine returns

following the SEO to determine whether any capital formation benefits are associated with

opportunistic or informative disclosure under SOR Greater stock returns prior to the issue with

no reversal afterward would be consistent with more informative disclosure and a richer

information environment during equity capital formation Greater stock returns prior to the issue

with a reversal afterward would be consistent with more opportunistic disclosure and a poorer

information environment during equity capital formation under SOR

24 Capital formation period

We define the capital formation period as the one-month (ie 30-day) period before the

SEO issue date in our tests for several reasons First it includes the SEO period when investors

assess the SEO price and decide to invest Prior literature indicates that firms using disclosure to

influence market expectations time the disclosure immediately preceding a pricing event (eg

Brockman et al 2008 Cheng and Lo 2006 Aboody and Kasznik 2000) Thus this period is

when disclosure is most likely to influence capital formation (eg Korajczyk et al 1991)

Second although most regulations are written with respect to the filing date disclosure was also

effectively limited before the issue date possibly as a result of cumbersome and uncertain

10

prospectus amendment requirements or firms viewing the prospectus amendment as a new

registration statement Ruland et al (1990) find that few firms issue equity within one month of a

forecast even though firms are likely to issue equity within three months of a forecast Marquardt

and Wiedman (1998) state that gun-jumping laws discourage forecasts in the period before an

SEO Third although the SEC notes in its comments on the quiet period that federal securities

laws do not define the term it uses a 30-day period prior to the filing date in SOR when

establishing bright-line disclosure restrictions for non-WKSIs

3 Data and sample selection

31 Disclosures analyzed

We study multiple forms of disclosure to investigate factual and forward-looking

information provided during the equity capital formation period We examine two types of

forward-looking disclosure termed ldquoguidancerdquo disclosure in this paper MEFs and forward-

looking FWPs MEFs are one of the most informative voluntary disclosures (Beyer et al 2010)

and we can directly measure accuracy for MEFs An annual (quarterly) MEF is a firm-issued

point or range forecast of annual (quarterly) earnings11

An FWP is a novel disclosure introduced

with SOR12

To broadly examine how SOR affects different types of firms we consider two additional

types of factual and forward-looking disclosure earnings announcements and 8-Ks13

This

11 Similar to Gong et al (2009) we include only point and range forecasts which constitute 94 of MEFs in the

First Call database during our sample period Forecast errors are not as clearly defined for other forms of forecasts

We use the mid-point of range forecasts to examine our accuracy research question and replace point forecasts with

the mid-point when the point forecast is outside the forecast range to correct First Call input errors 12

Although we focus on FWPs with forward-looking information firms may also file an FWP after issuing any

shelf-registered security to update the prospectus We exclude these FWPs because they summarize information

available in the final prospectus and are not forward-looking in nature 13

Examining 8-Ks allows us to look at a fuller set of disclosures 8 -Ks may include press releases of an earnings

announcement or MEF but they also typically contain additional information Our separate results for each of these

11

timing or both to affect the distance (ie number of days) under SOR Although firms have

specific deadlines for reporting items on an 8-K evidence by Segal and Segal (2013) shows that

managers engage in strategic timing Furthermore the results in Korajczyk et al (1991) suggest

that firms especially non-forecasting firms strategically time SEOs just after earnings

announcements because they prefer issuing equity when investors are most informed about firm

quality However firms may be less dependent upon earnings announcements (ie greater

distance) since they can provide additional disclosure prior to an SEO under SOR

Second we examine whether more MEFs within the capital formation period is

indicative of a richer information environment under SOR by investigating the accuracy of

MEFs Firms have dueling incentives to provide accurate MEFs prior to an SEO Firms may be

able to issue stock at a higher price following opportunistic disclosure (Lang and Lundholm

2000) However Li and Zhuang (2012) find that highly accurate MEFs reduce the cost of an

SEO MEF accuracy during the month prior to an SEO may also be subject to heightened

scrutiny and litigation risk despite the new safe harbor provisions Improved accuracy would be

consistent with informative voluntary disclosure during the capital formation period while worse

accuracy would be consistent with opportunistic voluntary disclosure as predicted by SORrsquos

critics (eg Morrissey 2007)

Finally we examine two aspects of the information environment reflected in stock prices

First we examine whether the magnitude of information measured as cumulative absolute

market-adjusted returns is greater during the capital formation period All else equal if

disclosure during the capital formation period is associated with greater information absolute

market-adjusted returns should be larger during the month prior to the SEO under SOR If a

larger information magnitude results from greater disclosure immediately preceding the issue

9

date then the information magnitude should become larger as the issue date approaches and for

firms providing disclosure within one week prior to the SEO Second we examine whether SOR

is associated with capital formation benefits by cumulating signed raw returns during the capital

formation period To the extent that SOR is associated with increased offering flexibility through

enhanced shelf registration and communication mechanisms SOR should be associated with

capital formation benefits reflected in higher cumulative stock returns during the capital

formation period ie a larger increase in stock price prior to the issue date We examine returns

following the SEO to determine whether any capital formation benefits are associated with

opportunistic or informative disclosure under SOR Greater stock returns prior to the issue with

no reversal afterward would be consistent with more informative disclosure and a richer

information environment during equity capital formation Greater stock returns prior to the issue

with a reversal afterward would be consistent with more opportunistic disclosure and a poorer

information environment during equity capital formation under SOR

24 Capital formation period

We define the capital formation period as the one-month (ie 30-day) period before the

SEO issue date in our tests for several reasons First it includes the SEO period when investors

assess the SEO price and decide to invest Prior literature indicates that firms using disclosure to

influence market expectations time the disclosure immediately preceding a pricing event (eg

Brockman et al 2008 Cheng and Lo 2006 Aboody and Kasznik 2000) Thus this period is

when disclosure is most likely to influence capital formation (eg Korajczyk et al 1991)

Second although most regulations are written with respect to the filing date disclosure was also

effectively limited before the issue date possibly as a result of cumbersome and uncertain

10

prospectus amendment requirements or firms viewing the prospectus amendment as a new

registration statement Ruland et al (1990) find that few firms issue equity within one month of a

forecast even though firms are likely to issue equity within three months of a forecast Marquardt

and Wiedman (1998) state that gun-jumping laws discourage forecasts in the period before an

SEO Third although the SEC notes in its comments on the quiet period that federal securities

laws do not define the term it uses a 30-day period prior to the filing date in SOR when

establishing bright-line disclosure restrictions for non-WKSIs

3 Data and sample selection

31 Disclosures analyzed

We study multiple forms of disclosure to investigate factual and forward-looking

information provided during the equity capital formation period We examine two types of

forward-looking disclosure termed ldquoguidancerdquo disclosure in this paper MEFs and forward-

looking FWPs MEFs are one of the most informative voluntary disclosures (Beyer et al 2010)

and we can directly measure accuracy for MEFs An annual (quarterly) MEF is a firm-issued

point or range forecast of annual (quarterly) earnings11

An FWP is a novel disclosure introduced

with SOR12

To broadly examine how SOR affects different types of firms we consider two additional

types of factual and forward-looking disclosure earnings announcements and 8-Ks13

This

11 Similar to Gong et al (2009) we include only point and range forecasts which constitute 94 of MEFs in the

First Call database during our sample period Forecast errors are not as clearly defined for other forms of forecasts

We use the mid-point of range forecasts to examine our accuracy research question and replace point forecasts with

the mid-point when the point forecast is outside the forecast range to correct First Call input errors 12

Although we focus on FWPs with forward-looking information firms may also file an FWP after issuing any

shelf-registered security to update the prospectus We exclude these FWPs because they summarize information

available in the final prospectus and are not forward-looking in nature 13

Examining 8-Ks allows us to look at a fuller set of disclosures 8 -Ks may include press releases of an earnings

announcement or MEF but they also typically contain additional information Our separate results for each of these

11

date then the information magnitude should become larger as the issue date approaches and for

firms providing disclosure within one week prior to the SEO Second we examine whether SOR

is associated with capital formation benefits by cumulating signed raw returns during the capital

formation period To the extent that SOR is associated with increased offering flexibility through

enhanced shelf registration and communication mechanisms SOR should be associated with

capital formation benefits reflected in higher cumulative stock returns during the capital

formation period ie a larger increase in stock price prior to the issue date We examine returns

following the SEO to determine whether any capital formation benefits are associated with

opportunistic or informative disclosure under SOR Greater stock returns prior to the issue with

no reversal afterward would be consistent with more informative disclosure and a richer

information environment during equity capital formation Greater stock returns prior to the issue

with a reversal afterward would be consistent with more opportunistic disclosure and a poorer

information environment during equity capital formation under SOR

24 Capital formation period

We define the capital formation period as the one-month (ie 30-day) period before the

SEO issue date in our tests for several reasons First it includes the SEO period when investors

assess the SEO price and decide to invest Prior literature indicates that firms using disclosure to

influence market expectations time the disclosure immediately preceding a pricing event (eg

Brockman et al 2008 Cheng and Lo 2006 Aboody and Kasznik 2000) Thus this period is

when disclosure is most likely to influence capital formation (eg Korajczyk et al 1991)

Second although most regulations are written with respect to the filing date disclosure was also

effectively limited before the issue date possibly as a result of cumbersome and uncertain

10

prospectus amendment requirements or firms viewing the prospectus amendment as a new

registration statement Ruland et al (1990) find that few firms issue equity within one month of a

forecast even though firms are likely to issue equity within three months of a forecast Marquardt

and Wiedman (1998) state that gun-jumping laws discourage forecasts in the period before an

SEO Third although the SEC notes in its comments on the quiet period that federal securities

laws do not define the term it uses a 30-day period prior to the filing date in SOR when

establishing bright-line disclosure restrictions for non-WKSIs

3 Data and sample selection

31 Disclosures analyzed

We study multiple forms of disclosure to investigate factual and forward-looking

information provided during the equity capital formation period We examine two types of

forward-looking disclosure termed ldquoguidancerdquo disclosure in this paper MEFs and forward-

looking FWPs MEFs are one of the most informative voluntary disclosures (Beyer et al 2010)

and we can directly measure accuracy for MEFs An annual (quarterly) MEF is a firm-issued

point or range forecast of annual (quarterly) earnings11

An FWP is a novel disclosure introduced

with SOR12

To broadly examine how SOR affects different types of firms we consider two additional

types of factual and forward-looking disclosure earnings announcements and 8-Ks13

This

11 Similar to Gong et al (2009) we include only point and range forecasts which constitute 94 of MEFs in the

First Call database during our sample period Forecast errors are not as clearly defined for other forms of forecasts

We use the mid-point of range forecasts to examine our accuracy research question and replace point forecasts with

the mid-point when the point forecast is outside the forecast range to correct First Call input errors 12

Although we focus on FWPs with forward-looking information firms may also file an FWP after issuing any

shelf-registered security to update the prospectus We exclude these FWPs because they summarize information

available in the final prospectus and are not forward-looking in nature 13

Examining 8-Ks allows us to look at a fuller set of disclosures 8 -Ks may include press releases of an earnings

announcement or MEF but they also typically contain additional information Our separate results for each of these

11

prospectus amendment requirements or firms viewing the prospectus amendment as a new

registration statement Ruland et al (1990) find that few firms issue equity within one month of a

forecast even though firms are likely to issue equity within three months of a forecast Marquardt

and Wiedman (1998) state that gun-jumping laws discourage forecasts in the period before an

SEO Third although the SEC notes in its comments on the quiet period that federal securities

laws do not define the term it uses a 30-day period prior to the filing date in SOR when

establishing bright-line disclosure restrictions for non-WKSIs

3 Data and sample selection

31 Disclosures analyzed

We study multiple forms of disclosure to investigate factual and forward-looking

information provided during the equity capital formation period We examine two types of

forward-looking disclosure termed ldquoguidancerdquo disclosure in this paper MEFs and forward-

looking FWPs MEFs are one of the most informative voluntary disclosures (Beyer et al 2010)

and we can directly measure accuracy for MEFs An annual (quarterly) MEF is a firm-issued

point or range forecast of annual (quarterly) earnings11

An FWP is a novel disclosure introduced

with SOR12

To broadly examine how SOR affects different types of firms we consider two additional

types of factual and forward-looking disclosure earnings announcements and 8-Ks13

This

11 Similar to Gong et al (2009) we include only point and range forecasts which constitute 94 of MEFs in the

First Call database during our sample period Forecast errors are not as clearly defined for other forms of forecasts

We use the mid-point of range forecasts to examine our accuracy research question and replace point forecasts with

the mid-point when the point forecast is outside the forecast range to correct First Call input errors 12

Although we focus on FWPs with forward-looking information firms may also file an FWP after issuing any

shelf-registered security to update the prospectus We exclude these FWPs because they summarize information

available in the final prospectus and are not forward-looking in nature 13

Examining 8-Ks allows us to look at a fuller set of disclosures 8 -Ks may include press releases of an earnings

announcement or MEF but they also typically contain additional information Our separate results for each of these

11

approach allows us to study non-forecasting firms with a policy of providing disclosure besides

MEFs and more thoroughly examine other forms of disclosure by forecasting firms Firms have

less discretion over whether and when to provide earnings announcements and 8-Ks than MEFs

but may provide both factual and forward-looking information in these disclosures Earnings

announcements and 8-Ks are two of the primary mediums for ongoing and periodic disclosure

for publicly listed firms The earnings announcement date is a particularly important event for

resolving information asymmetry prior to an SEO (Korajczyk et al 1991) 8-Ks are intended to

maintain the accuracy and adequacy of information disclosed by firms by providing a continuous

stream of disclosure to investors around significant or material corporate events They constitute

more than half of all SEC filings by public firms and are associated with informative disclosure

that are complementary to periodic reports (Lerman and Livnat 2010 Drake et al 2014) In

addition the size of information in 8-Ks can be directly measured Hanley and Hoberg (2013)

find the size of information releases prior to equity offerings impacts the cost of capital For our

analysis we measure the size of the last 8-K filed prior to the SEO issue date14

To control for

potential bias created by SEC changes to 8-K disclosure requirements in 2004 we also measure

8-K abnormal size as the SEO firm 8-K size less the median non-SEO WKSI 8-K size during the

same period15

types of disclosures indicate that our 8-K results are not driven by the inclusion of earnings announcements and

MEFs in the 8-K sample Few firms provide an MEF and we see no shift in the earnings announcements 14

We expect the 8-K file size to be highly correlated with the amount of information provided Other studies use

word counts to proxy for disclosure size (eg Francis et al 2002) For robustness we conduct a word count on 30

randomly selected 8-Ks around SOR File size is 86 correlated with the number of words in both periods 15

The SEC changed the 8-K disclosure requirements effective August 23 2004 [see Lerman and Livnat (2010) for a

more detailed discussion of these changes] which may impact the size of the 8-K independently from SOR To be

considered a non-SEO WKSI for our additional 8-K size measurements a firm must meet the WKSI criteria for

market capitalization ($700 million) and must not conduct an SEO during the same calendar year as the sa mple firm

12

32 Sample selection

We obtain data on SEOs for January 1 2002 through December 31 2009 from the

Securities Data Company (SDC) Global New Issues database MEFs and actual earnings are

from First Call16

8-K and FWP disclosures are from the Wharton Research Data Services

(WRDS) SEC Analytics Suite and from the Electronic Data-Gathering Analysis and Retrieval

(EDGAR) system Earnings announcement dates and financial data are from Compustat Stock

return data are from the Center for Research in Security Prices (CRSP) and institutional

ownership data are from the Thomson Reuters 13F Holdings Database Analyst earnings

forecasts are from the Institutional Brokersrsquo Estimate System (IBES)

We verify offering details through a firmrsquos registration statement preliminary prospectus

and final prospectus all of which are collected from EDGAR We identify WKSIs as SEO firms

that have been publicly traded for at least one year and whose market value of equity for the

calendar year-end before the offering is at least $700 million These criteria correctly identify

over 99 of firms that file as WKSIs after SOR We exclude pure secondary offerings (in which

100 of the shares are sold by individuals in the secondary market and the firm raises no

capital) rights offerings offerings by regulated firms (ie financials and utilities) due to their

different disclosure requirements and offerings by non-US firms Finally we exclude 27 firms

with missing data from CRSP or Compustat and three firms without coverage in First Call17

The

final sample contains 360 SEOs issued by 243 unique firms18

We designate SEOs with issue

dates from January 1 2002 to November 30 2005 as pre-SOR and SEOs with issue dates from

16 Our focus on WKSIs after 2001 reduces the potential First Call bias identified by Chuk et al (2013) They find the

number of MEFs using First Call may be biased downward when examining earlier time periods and firms that are

small and not as widely followed but they find little evidence of bias after the 1990s and for firms that are large or

more widely followed 17

We exclude one observation in which the stock price is less than $100 at the beginning of the fiscal year 18

Although only 25 WKSIs issue equity both before and after SOR we repeat our analysis for this subsample and

find qualitatively similar results

13

December 1 2005 through December 31 2009 as post-SOR providing a four-year period on

either side of SOR19

4 Research design

41 Occurrence frequency 8-K size and distance

For our first test examining whether more factual and forward-looking disclosure exists

during the capital formation period under SOR we specify the following equation

Y = α + β1 SOR + β2 Ln(market capitalization) + β3 Institutional ownership

+ β4 Residual analyst following + β5 Analyst dispersion + β6 Litigation risk

+ β7 Book-to-market + β8 Loss + β9 Firm risk + ε (1)

We estimate Eq (1) for each of our dependent variables (Y) which are measures of

occurrence frequency 8-K size and distance We employ a probit regression to estimate Eq (1)

for occurrence because ordinary least squares (OLS) is an inefficient estimation technique when

the dependent variable is binary rather than continuous (Aldrich and Nelson 1984) We estimate

Eq (1) for frequency size and distance using OLS20

We examine occurrence for SEOs when

the firm provides an annual MEF as well as for those that provide any guidance disclosure We

study frequency before an SEO for the number of guidance disclosures 8-Ks and total

disclosures We consider 8-K size for the last 8-K within the capital formation period Finally

we examine distance which is the number of days between the last disclosure (guidance 8-K

and any of the disclosures we capture) and the SEO issue date Our variable of interest SOR is

equal to one for observations in which the SEO issue date is on or after December 1 2005 We

19 Six SEOs are issued in December 2005 and are included in the post-SOR sample

20 We also estimate Eq (1) for frequency using an OLS regression with the natural log of frequency a Tobit

regression a Poisson regression and a negative binomial regression Our results are robust to each alternative specification

14

include control variables found to be associated with disclosure21

Full definitions are presented

in the Appendix With the exception of fiscal period end earnings all measures are calculated at

least 30 days prior to the issue date We use robust standard errors clustered at the firm level

throughout our analysis

42 Management earnings forecast accuracy

Our second test examines the richness of the information environment associated with

differences in MEF disclosure under SOR by measuring MEF accuracy Focusing on MEFs

provided during the month prior to the SEO issue date we estimate accuracy as follows We

measure MEF error as the absolute value of the difference and bias as the signed difference

between the firmrsquos last earnings per share forecast (annual or annualized quarterly) and the actual

earnings per share scaled by the stock price at prior fiscal year-end22

Similar to Ajinkya et al

(2005) we use actual earnings data from First Call to ensure consistency with MEFs In addition

First Callrsquos earnings data reflect street (or core) earnings which are of higher quality than GAAP

earnings for valuation purposes (Gu and Chen 2004)

We test MEF accuracy controlling for other factors that may influence forecast accuracy

using the following equation

Z = α + β1 SOR + β2 Ln(market capitalization) + β3 Institutional ownership

+ β4 Residual analyst following + β5 Analyst dispersion + β6 Litigation risk

+ β7 Book-to-market + β8 Loss + β9 Firm risk + β10 Horizon + ε (2)

21 Prior research (eg Ajinkya et al 2005) often includes major auditor as an additional control variable The

biggest four audit firms audit 96 of our sample As a result we exclude this control variable Because the number

of analysts following a firm is highly correlated with firm size we regress analyst following on Ln(market

capitalization) and retain the residuals as residual analyst following This allows us to control for the component of

analyst following not explained by firm size (Chang et al 2006) We use analyst following as a robustness check

and obtain similar results (not tabulated) 22

We repeat the analysis scaling by earnings per share and obtain similar results (not tabulated)

15

include control variables found to be associated with disclosure21

Full definitions are presented

in the Appendix With the exception of fiscal period end earnings all measures are calculated at

least 30 days prior to the issue date We use robust standard errors clustered at the firm level

throughout our analysis

42 Management earnings forecast accuracy

Our second test examines the richness of the information environment associated with

differences in MEF disclosure under SOR by measuring MEF accuracy Focusing on MEFs

provided during the month prior to the SEO issue date we estimate accuracy as follows We

measure MEF error as the absolute value of the difference and bias as the signed difference

between the firmrsquos last earnings per share forecast (annual or annualized quarterly) and the actual

earnings per share scaled by the stock price at prior fiscal year-end22

Similar to Ajinkya et al

(2005) we use actual earnings data from First Call to ensure consistency with MEFs In addition

First Callrsquos earnings data reflect street (or core) earnings which are of higher quality than GAAP

earnings for valuation purposes (Gu and Chen 2004)

We test MEF accuracy controlling for other factors that may influence forecast accuracy

using the following equation

Z = α + β1 SOR + β2 Ln(market capitalization) + β3 Institutional ownership

+ β4 Residual analyst following + β5 Analyst dispersion + β6 Litigation risk

+ β7 Book-to-market + β8 Loss + β9 Firm risk + β10 Horizon + ε (2)

21 Prior research (eg Ajinkya et al 2005) often includes major auditor as an additional control variable The

biggest four audit firms audit 96 of our sample As a result we exclude this control variable Because the number

of analysts following a firm is highly correlated with firm size we regress analyst following on Ln(market

capitalization) and retain the residuals as residual analyst following This allows us to control for the component of

analyst following not explained by firm size (Chang et al 2006) We use analyst following as a robustness check

and obtain similar results (not tabulated) 22

We repeat the analysis scaling by earnings per share and obtain similar results (not tabulated)

15

We estimate Eq (2) for both of our accuracy dependent variables (Z) error and bias

which are measured for the last MEF for firms providing an MEF during the specified period

prior to the SEO issue date Similar to Eq (1) our variable of interest SOR is equal to one if the

issue date is on or after December 1 2005

43 Information magnitude and stock returns

For our third test we examine stock returns in two ways to provide additional analysis of

the information environment during equity capital formation under SOR First we measure the

magnitude of information released with disclosure by analyzing the absolute value of market-

adjusted returns similar to previous studies (eg Francis et al 2002 Bailey et al 2006)

Specifically we measure magnitude as the cumulative absolute daily returns net of the return on

the CRSP value-weighted index over the [-200] trading day period to proxy for the magnitude of

information reflected in stock prices during the capital formation period with day 0 being the

volume-based corrected issue date23

This 21-trading-day window closely corresponds to the

one-month disclosure window used in our disclosure tests Thus magnitude [-200] captures all

information reflected in stock prices (Tetlock 2010) during the capital formation period

including information conveyed by the disclosures we examine Although Henry and Koski

(2010) find no evidence of manipulative short selling prior to shelf-registered SEOs presumably

because anticipating when these SEOs will occur is difficult we focus on abnormal information

magnitude to control for factors other than capital formation disclosure such as increased trading

23 For analysis of returns during the period surrounding an SEO we employ a volume -based correction methodology

because late day or overnight deals may result in processing the trading volume on the following trading day (ie

the day after the issue date in SDC) Following Corwin (2003) we adjust the issue day later by one day when the

trading volume on the day after the issue date in SDC is more than double the trading volume on the SDC issue date

and when it is more than two times the mean trading volume over the past 250 trading days The issue day is

corrected for 65 of our sample

16

volume that may influence magnitude during the period surrounding an SEO We define

abnormal magnitude as magnitude [-200] less magnitude [+1+20]

Second we examine signed stock returns to explore whether SOR is associated with

capital formation benefits and more informative or opportunistic disclosure We cumulate the

signed raw daily stock returns (return) during the [-200] trading day period to measure how

information is incorporated into the stock price during the capital formation period Because

SOR impacts the shelf registration process in addition to permissible disclosure return is likely

to be a lower bound for the net effect of increased disclosure under SOR as the information

content of a shelf registration or SEO announcement shifts from the filing date to the pre-issue

date Thus a negative SEO announcement return is more likely to be included in return under

SOR Similar to abnormal magnitude we measure capital formation benefits through abnormal

return defined as return [-200] less return [+1+20] We further examine the period after the

SEO issue date return [+1+20] to distinguish between informative and opportunistic

disclosure A greater return during the capital formation period followed by a reversal after the

issue date post-SOR would be consistent with opportunistic disclosure under SOR Conversely a

greater return with no reversal afterward would be consistent with informative disclosure24

44 Comparison with other studies

Our research design differs from other studies in important ways First we use the issue

date as the SEO date in response to changes in the SEO environment Earlier studies on pre-SEO

disclosure (eg Marquardt and Wiedman 1998 Lang and Lundholm 2000) refer to the filing

or registration date as the SEO announcement date because their samples are composed of non-

24 To help ensure that our post-SOR period captures any potential reversal due to opportunistic disclosure we extend

the post-SOR period to +60 trading days ie one additional quarter Results are robust

17

volume that may influence magnitude during the period surrounding an SEO We define

abnormal magnitude as magnitude [-200] less magnitude [+1+20]

Second we examine signed stock returns to explore whether SOR is associated with

capital formation benefits and more informative or opportunistic disclosure We cumulate the

signed raw daily stock returns (return) during the [-200] trading day period to measure how

information is incorporated into the stock price during the capital formation period Because

SOR impacts the shelf registration process in addition to permissible disclosure return is likely

to be a lower bound for the net effect of increased disclosure under SOR as the information

content of a shelf registration or SEO announcement shifts from the filing date to the pre-issue

date Thus a negative SEO announcement return is more likely to be included in return under

SOR Similar to abnormal magnitude we measure capital formation benefits through abnormal

return defined as return [-200] less return [+1+20] We further examine the period after the

SEO issue date return [+1+20] to distinguish between informative and opportunistic

disclosure A greater return during the capital formation period followed by a reversal after the

issue date post-SOR would be consistent with opportunistic disclosure under SOR Conversely a

greater return with no reversal afterward would be consistent with informative disclosure24

44 Comparison with other studies

Our research design differs from other studies in important ways First we use the issue

date as the SEO date in response to changes in the SEO environment Earlier studies on pre-SEO

disclosure (eg Marquardt and Wiedman 1998 Lang and Lundholm 2000) refer to the filing

or registration date as the SEO announcement date because their samples are composed of non-

24 To help ensure that our post-SOR period captures any potential reversal due to opportunistic disclosure we extend

the post-SOR period to +60 trading days ie one additional quarter Results are robust

17

shelf (traditional) offerings In non-shelf offerings firms convey information about the upcoming

SEO on the registration date and the issue date usually occurs soon after registration (Bethel and

Krigman 2008) In contrast over 80 of our sample issues equity from a shelf registration and

the filing date predates the issue date by 257 days on average (Table 1)25

Second we examine

numerous properties of disclosure during the one-month period prior to the SEO issue date In

addition to MEFs and earnings releases we study other forms of disclosure namely 8-Ks and

FWPs To our knowledge this study is the first to examine these forms of disclosure and the first

to examine the accuracy of MEFs provided during the capital formation period Finally we

analyze the forecasting and non-forecasting samples separately to provide insight into pre-SEO

disclosure activity for firms with and firms without a recent history of forecasting Li and

Zhuang (2012) examine management forecasts primarily in the form of MEFs during the 24-

and 12-month period prior to the filing date for SEOs in a more recent period 1997 to 2006

They find that management forecasts are associated with less underpricing in SEOs by smaller

firms but that no significant relation exists in SEOs by larger firms The insignificant results for

large firms may result from either the potential misidentification of the capital formation period

or the influence of both guidance and non-guidance disclosure on the information environment

Shroff et al (2013) examine disclosure before and after SOR However their focus is more

similar to Lang and Lundholm (2000) and Li and Zhuang (2012) Shroff et al focus on the pre-

filing period include two forms of disclosure (forecasts and press releases) and use a restricted

25 Henry and Koski (2010) and Autore et al (2008) discuss the importance and ch allenges of identifying

announcement dates for SEOs especially for shelf registrations Consistent with these studies we find no significant

market reaction at the filing of shelf registrations The median three-day abnormal pre-issue return is -32 which

is consistent with estimates of the announcement returns for shelf-registered SEOs during previous periods (eg

Autore et al 2008) Because most SEOs are now shelf-registered future researchers will want to focus on the pre-

issue date or issue date and not the filing date as the appropriate information event

18

sample with predominantly smaller firms which remain subject to some disclosure restrictions

under SOR

5 Results

51 Descriptive statistics

Table 1 Panel A presents the distribution of SEOs according to the issuing firmrsquos MEF

policy We classify SEOs according to whether the firm has chosen to provide at least one annual

or quarterly point or range MEF or both within one year before the issue date (forecasting

sample) or whether the issuing firm has chosen not to provide an MEF (non-forecasting

26 27sample) Managers issue MEFs within one year of 126 SEOs or 35 of the sample Thirty-

eight percent (33) of SEOs are in the forecasting sample pre-SOR (post-SOR) This is

consistent with firms maintaining an established forecasting policy under SOR Thus greater use

of MEFs during the capital formation period under SOR is likely to be confined to firms with a

previous regular release of MEFs (ie forecasting firms)

[Insert Table 1 near here]

Panel B presents SEO characteristics The percentage of shelf-registered SEOs by WKSIs

increases during our sample period from 75 pre-SOR to 87 post-SOR28

Moreover the

average number of days between the filing and issue dates increases from 196 days pre-SOR to

298 days post-SOR In fact the filing date predates the issue date by at least 90 days for 55 of

the SEOs in our sample indicating that over half our sample would be excluded using

26 We find in our sample that forecasting disclosure policy tends to be static over time This approach therefore

allows for a reasonable time period to identify a firmrsquos existing MEF policy and minimizes any bias resulting from limited coverage in First Call prior to our sample period 27

This proportion is consistent with other studies (eg Beyer et al 2010 Chuk et al 2013) 28

We spoke with a chief financial officer of a Fortune 500 company who indicated that with the ease of shelf registration his company employs a non-shelf offering only when the use of proceeds falls outside ordinary use

19

sample with predominantly smaller firms which remain subject to some disclosure restrictions

under SOR

5 Results

51 Descriptive statistics

Table 1 Panel A presents the distribution of SEOs according to the issuing firmrsquos MEF

policy We classify SEOs according to whether the firm has chosen to provide at least one annual

or quarterly point or range MEF or both within one year before the issue date (forecasting

sample) or whether the issuing firm has chosen not to provide an MEF (non-forecasting

26 27sample) Managers issue MEFs within one year of 126 SEOs or 35 of the sample Thirty-

eight percent (33) of SEOs are in the forecasting sample pre-SOR (post-SOR) This is

consistent with firms maintaining an established forecasting policy under SOR Thus greater use

of MEFs during the capital formation period under SOR is likely to be confined to firms with a

previous regular release of MEFs (ie forecasting firms)

[Insert Table 1 near here]

Panel B presents SEO characteristics The percentage of shelf-registered SEOs by WKSIs

increases during our sample period from 75 pre-SOR to 87 post-SOR28

Moreover the

average number of days between the filing and issue dates increases from 196 days pre-SOR to

298 days post-SOR In fact the filing date predates the issue date by at least 90 days for 55 of

the SEOs in our sample indicating that over half our sample would be excluded using

26 We find in our sample that forecasting disclosure policy tends to be static over time This approach therefore

allows for a reasonable time period to identify a firmrsquos existing MEF policy and minimizes any bias resulting from limited coverage in First Call prior to our sample period 27

This proportion is consistent with other studies (eg Beyer et al 2010 Chuk et al 2013) 28

We spoke with a chief financial officer of a Fortune 500 company who indicated that with the ease of shelf registration his company employs a non-shelf offering only when the use of proceeds falls outside ordinary use

19

restrictions commonly imposed in studies focusing on the filing date (eg Lang and Lundholm

2000 Shroff et al 2013)29

The average number of days between the filing of a preliminary

prospectus (pre-issue date) and the issue date drops from three to two Thus information

provided at the shelf registration is more likely to be dated and underwriters could have less

time to perform sufficient due diligence when the equity is issued post-SOR (Blackwell et al

1990 Coffee 2005) The average proceeds raised increases from $287 million pre-SOR to $394

million post-SOR However the average SEO size represents just over 12 of the SEO firmrsquos

market value of equity in both periods Panel C presents the average annual number of 8-Ks filed

by WKSIs issuing equity during the fiscal year Non-forecasting firms appear to rely more

heavily upon 8-Ks throughout the year than the forecasting firms but the number of 8-Ks is

greater for both groups over time

Table 2 presents descriptive statistics for our control variables Mean values are presented

for the pre- and post-SOR full forecasting and non-forecasting samples The average market

capitalization analyst following and percentage of firms reporting a loss at fiscal year-end are

similar in both periods Institutional ownership and firm risk are greater in all post-SOR samples

However the proportion of observations in high litigation risk industries is similar in the

forecasting sample and lower in the non-forecasting sample post-SOR The non-forecasting

sample is also made up of smaller firms that are more likely to report a loss have slightly greater

analyst following and have greater analyst forecast dispersion post-SOR

[Insert Table 2 near here]

29 The increased distance between the filing and issue dates is not confined to WKSIs For example in SDC the

filing date predates the issue date during our sample period by at least 90 days for 40 of SEOs by seasoned issuers

firms that are smaller than WKSIs and receive fewer benefits under SOR

20

52 Results for occurrence frequency 8-K size and distance

Our first test investigates the various ways firms may use relaxed disclosure restrictions

under SOR to provide more information in the capital formation period Although we do not

expect disclosure to decrease under SOR we use a conservative two-tailed test to present

consistent statistics throughout the paper Table 3 presents univariate results for occurrence

(Panel A) frequency (Panel B) 8-K size (Panel C) and distance (Panel D) Occurrence

frequency and 8-K size are each set equal to zero when no disclosure occurs during the one-

month period prior to the issue date30

No greater occurrence of MEFs is found during the capital

formation period in the post-SOR full sample However annual MEFs are significantly more

likely to occur in the post-SOR forecasting sample relative to the pre-SOR forecasting sample

(451 versus 273) and 127 of the post-SOR forecasting sample provide an FWP For the

post-SOR non-forecasting sample 76 file an FWP during the month before an SEO This may

indicate that firms not typically providing MEFs choose to provide guidance disclosure via an

FWP under SOR as opposed to changing forecasting policy

[Insert Table 3 near here]

Total disclosure frequency is 25 larger post-SOR relative to pre-SOR (308 versus

246) Total disclosure consists of MEFs FWPs 8-Ks and earnings announcements For the

post-SOR forecasting sample higher frequency of total disclosure (358 versus 271) is driven by

the higher frequency of annual MEFs (059 versus 031) and 8-Ks (192 versus 158) and the use

30 Our univariate and multivariate results for frequency and 8-K size are robust to excluding firms without disclosure

during the one-month period prior to the issue date (ie excluding zeroes)

21

of FWPs31

For the post-SOR non-forecasting sample total disclosure frequency is greater (283

versus 231) as a result of more frequent 8-Ks (223 versus 182) as well as the use of FWPs

8-K size and 8-K abnormal size are significantly greater post-SOR relative to pre-SOR

(597 versus 254 and 517 versus 221 respectively) The differences are also economically

significant in the forecasting and non-forecasting samples However the statistical significance

weakens in the smaller samples The difference in size is significant in both samples in a one-

tailed test but is only significant in the forecasting sample in a more conservative two-tailed test

Although distance cannot be measured for FWPs pre-SOR they have the shortest

distance during the post-SOR capital formation period averaging 315 days The distance

between the last disclosure of any type during the capital formation period and the issue date is

significantly shorter post-SOR (615 versus 785 days) The shorter distance is primarily found

for guidance (ie MEF or FWP) in the post-SOR forecasting sample and for 8-Ks in the post-

SOR non-forecasting sample The distance during the capital formation period is 485 days

shorter in the post-SOR forecasting sample for last guidance and is 178 days shorter in the post-

SOR non-forecasting sample for last 8-K

Overall the results in Table 3 indicate that SEO firms are providing more information

during the capital formation period under SOR We find that a higher proportion of SEOs in the

post-SOR forecasting sample is preceded by an annual MEF in the month before the issue date

and we find SEOs in both the forecasting and non-forecasting samples are preceded by FWPs

post-SOR Frequency and size of 8-Ks are greater in the post-SOR full sample Pre-SEO

disclosure is also closer before the issue date However the shorter distance results from closer

MEFs in the forecasting sample and closer 8-Ks in the non-forecasting sample post-SOR

31 In untabulated results we verify that the MEF results are not due to general trends in forecasting earnings In

contrast to our results we find the average annual frequency of quarterly MEFs decreases by 27 and the average

annual frequency of annual MEFs remains fairly constant in First Call from the pre- to post-SOR period

22

The multivariate results in Panel A of Table 4 are consistent with the univariate results

for occurrence frequency and 8-K size The significantly positive coefficient on SOR for the

forecasting sample in Columns 1ndash2 indicate that the likelihood of providing an annual MEF (p =

009) as well as guidance (p = 006) during the capital formation period is greater under SOR32

In addition Column 3 indicates guidance frequency in the capital formation period is greater

under SOR (p = 004) Overall the sign of the coefficients on the control variables loading

significantly are consistent with previous studies The significantly positive coefficients on SOR

(p = 001) in Columns 4ndash5 for the full sample show that both 8-Ks and total disclosures are more

frequent during the capital formation period under SOR The significantly positive coefficients

on SOR in Columns 6ndash7 for the full sample indicate that 8-K size of the last 8-K preceding the

SEO during the capital formation period is significantly larger under SOR even when

controlling for median size during the same period for WKSIs not issuing equity

[Insert Table 4 near here]

The multivariate results in Panel B provide additional evidence of shorter distance under

SOR for the last disclosure prior to the SEO issue date Columns 1ndash3 present results for last

disclosure restricted to the capital formation period The distance for last guidance is

significantly shorter (-563 days p = 003) and distance for last disclosure of any type is weakly

shorter (-106 days p = 020) under SOR The results in Columns 1ndash3 only partially capture the

relation between distance and SOR because the capital formation period restriction excludes the

last disclosure for a portion of the sample and does not account for the disproportionate

representation of the post-SOR sample Columns 4ndash6 remove the capital formation period

32 In untabulated results for the full sample the coefficient on SOR is not significant in a two-tailed test when

occurrence refers to an annual MEF (p = 017) but becomes significant when occurrence refers to an annual MEF or

FWP (p = 001) Thus greater guidance disclosure in the full sample during the capital formation period under SOR

stems from the use of FWPs and greater use of annual MEFs during the month prior to the SEO issue date

23

restriction and present results for last disclosure within one year prior to the issue date When

removing the restriction distance is significantly shorter for last 8-K (-691 days p = 002) and

last disclosure of any type (-372 days p = 004) under SOR Overall the results in Table 4

support the univariate results and indicate that SEO firms are providing more shorter distance

disclosure and thus more information during the month before the SEO issue date 33

53 Results for management earnings forecast accuracy

Table 5 presents univariate results for our second test Panel A presents error and bias

results for the last MEF (annual or annualized quarterly) within the one-month period prior to an

SEO to see if MEFs most likely to influence capital formation are more or less accurate post-

SOR34

The average forecast error is 70 lower under SOR (072 versus 243 p lt 001) In

addition average forecast bias is 83 smaller (133 versus 023 p = 009)

[Insert Table 5 near here]

To further explore the source of greater forecast accuracy Panel B (C) presents the

accuracy results for the first and the last annual (quarterly) MEF along with forecast horizons

Here first annual (quarterly) MEF is the first annual (quarterly) MEF provided during the one

year prior to the SEO issue date and last annual (quarterly) MEF is the last annual (quarterly)

MEF provided during the month before the SEO issue date We also present the difference in the

first and last annual (quarterly) MEF before and after SOR to see if firms are increasing or

decreasing their forecast accuracy as the SEO issue date approaches

33 In untabulated results we re-estimate all of our equations including additional measures that proxy for litigation

risk (Kim and Skinner 2012) to help ensure the SOR variable does not proxy for increased litigation risk Our

results for SOR remain robust when we include these additional measures Results are also robust to the inclusion of

SOR interacted with litigation risk The interaction is insignificant in the re-estimation of Table 4 34

We use annual MEF measures if a firm provides both annual and quarterly MEFs during the one-month period

prior to an SEO

24

The results indicate that greater accuracy for the last MEF under SOR is primarily driven

by increasing accuracy from first to last annual MEF and by maintaining as opposed to

decreasing accuracy from first to last quarterly MEF as the issue date approaches For example

error drops significantly (p = 001) from the first (136) to the last (077) annual MEF post-

SOR but not pre-SOR Likewise bias decreases significantly from 071 to 032 post-SOR

but not pre-SOR Thus while accuracy is somewhat greater for the first annual MEF under SOR

the significant improvement occurs with the last annual MEF within one month prior to the SEO

issue date Moreover the last annual MEF averages 175 days prior to the fiscal year-end post-

SOR and is not significantly closer to the fiscal year-end than the last annual MEF pre-SOR

With quarterly MEFs error and bias for the last forecast post-SOR are also significantly lower

than for the last forecast pre-SOR However quarterly MEF accuracy is similar for the first and

the last quarterly MEFs post-SOR Conversely accuracy appears to deteriorate between the first

and the last quarterly MEF pre-SOR For example pre-SOR error (bias) increases from 136

(051) for the first to 234 (106) for the last quarterly MEF The pre-SOR increase in error

(098) is significantly greater than the post-SOR decrease (-009) at the 10 level Horizon

is similar for the pre- and post-SOR last quarterly MEF While greater accuracy does not appear

to be attributed to shorter last MEF horizon we control for horizon in a multivariate setting35

Table 6 presents the multivariate results for MEF accuracy While we focus on accuracy

of last MEFs during the capital formation period in Columns 1 and 3 we also include last MEFs

that are not updated in the capital formation period in Columns 2 and 4 to determine if results are

similar We find SOR is associated with significantly lower error and bias for the last MEF

during the one month and one year prior to the SEO issue date (p = 003 and 003 for error p =

35 In untabulated results we perform the same analysis without restricting the last forecast to the prior month before

the SEO We find similar results for annual MEFs However we find no greater accuracy for quarterly MEFs when

they are not restricted to the prior one-month period

25

of FWPs31

For the post-SOR non-forecasting sample total disclosure frequency is greater (283

versus 231) as a result of more frequent 8-Ks (223 versus 182) as well as the use of FWPs

8-K size and 8-K abnormal size are significantly greater post-SOR relative to pre-SOR

(597 versus 254 and 517 versus 221 respectively) The differences are also economically

significant in the forecasting and non-forecasting samples However the statistical significance

weakens in the smaller samples The difference in size is significant in both samples in a one-

tailed test but is only significant in the forecasting sample in a more conservative two-tailed test

Although distance cannot be measured for FWPs pre-SOR they have the shortest

distance during the post-SOR capital formation period averaging 315 days The distance

between the last disclosure of any type during the capital formation period and the issue date is

significantly shorter post-SOR (615 versus 785 days) The shorter distance is primarily found

for guidance (ie MEF or FWP) in the post-SOR forecasting sample and for 8-Ks in the post-

SOR non-forecasting sample The distance during the capital formation period is 485 days

shorter in the post-SOR forecasting sample for last guidance and is 178 days shorter in the post-

SOR non-forecasting sample for last 8-K

Overall the results in Table 3 indicate that SEO firms are providing more information

during the capital formation period under SOR We find that a higher proportion of SEOs in the

post-SOR forecasting sample is preceded by an annual MEF in the month before the issue date

and we find SEOs in both the forecasting and non-forecasting samples are preceded by FWPs

post-SOR Frequency and size of 8-Ks are greater in the post-SOR full sample Pre-SEO

disclosure is also closer before the issue date However the shorter distance results from closer

MEFs in the forecasting sample and closer 8-Ks in the non-forecasting sample post-SOR

31 In untabulated results we verify that the MEF results are not due to general trends in forecasting earnings In

contrast to our results we find the average annual frequency of quarterly MEFs decreases by 27 and the average

annual frequency of annual MEFs remains fairly constant in First Call from the pre- to post-SOR period

22

The multivariate results in Panel A of Table 4 are consistent with the univariate results

for occurrence frequency and 8-K size The significantly positive coefficient on SOR for the

forecasting sample in Columns 1ndash2 indicate that the likelihood of providing an annual MEF (p =

009) as well as guidance (p = 006) during the capital formation period is greater under SOR32

In addition Column 3 indicates guidance frequency in the capital formation period is greater

under SOR (p = 004) Overall the sign of the coefficients on the control variables loading

significantly are consistent with previous studies The significantly positive coefficients on SOR

(p = 001) in Columns 4ndash5 for the full sample show that both 8-Ks and total disclosures are more

frequent during the capital formation period under SOR The significantly positive coefficients

on SOR in Columns 6ndash7 for the full sample indicate that 8-K size of the last 8-K preceding the

SEO during the capital formation period is significantly larger under SOR even when

controlling for median size during the same period for WKSIs not issuing equity

[Insert Table 4 near here]

The multivariate results in Panel B provide additional evidence of shorter distance under

SOR for the last disclosure prior to the SEO issue date Columns 1ndash3 present results for last

disclosure restricted to the capital formation period The distance for last guidance is

significantly shorter (-563 days p = 003) and distance for last disclosure of any type is weakly

shorter (-106 days p = 020) under SOR The results in Columns 1ndash3 only partially capture the

relation between distance and SOR because the capital formation period restriction excludes the

last disclosure for a portion of the sample and does not account for the disproportionate

representation of the post-SOR sample Columns 4ndash6 remove the capital formation period

32 In untabulated results for the full sample the coefficient on SOR is not significant in a two-tailed test when

occurrence refers to an annual MEF (p = 017) but becomes significant when occurrence refers to an annual MEF or

FWP (p = 001) Thus greater guidance disclosure in the full sample during the capital formation period under SOR

stems from the use of FWPs and greater use of annual MEFs during the month prior to the SEO issue date

23

restriction and present results for last disclosure within one year prior to the issue date When

removing the restriction distance is significantly shorter for last 8-K (-691 days p = 002) and

last disclosure of any type (-372 days p = 004) under SOR Overall the results in Table 4

support the univariate results and indicate that SEO firms are providing more shorter distance

disclosure and thus more information during the month before the SEO issue date 33

53 Results for management earnings forecast accuracy

Table 5 presents univariate results for our second test Panel A presents error and bias

results for the last MEF (annual or annualized quarterly) within the one-month period prior to an

SEO to see if MEFs most likely to influence capital formation are more or less accurate post-

SOR34

The average forecast error is 70 lower under SOR (072 versus 243 p lt 001) In

addition average forecast bias is 83 smaller (133 versus 023 p = 009)

[Insert Table 5 near here]

To further explore the source of greater forecast accuracy Panel B (C) presents the

accuracy results for the first and the last annual (quarterly) MEF along with forecast horizons

Here first annual (quarterly) MEF is the first annual (quarterly) MEF provided during the one

year prior to the SEO issue date and last annual (quarterly) MEF is the last annual (quarterly)

MEF provided during the month before the SEO issue date We also present the difference in the

first and last annual (quarterly) MEF before and after SOR to see if firms are increasing or

decreasing their forecast accuracy as the SEO issue date approaches

33 In untabulated results we re-estimate all of our equations including additional measures that proxy for litigation

risk (Kim and Skinner 2012) to help ensure the SOR variable does not proxy for increased litigation risk Our

results for SOR remain robust when we include these additional measures Results are also robust to the inclusion of

SOR interacted with litigation risk The interaction is insignificant in the re-estimation of Table 4 34

We use annual MEF measures if a firm provides both annual and quarterly MEFs during the one-month period

prior to an SEO

24

The results indicate that greater accuracy for the last MEF under SOR is primarily driven

by increasing accuracy from first to last annual MEF and by maintaining as opposed to

decreasing accuracy from first to last quarterly MEF as the issue date approaches For example

error drops significantly (p = 001) from the first (136) to the last (077) annual MEF post-

SOR but not pre-SOR Likewise bias decreases significantly from 071 to 032 post-SOR

but not pre-SOR Thus while accuracy is somewhat greater for the first annual MEF under SOR

the significant improvement occurs with the last annual MEF within one month prior to the SEO

issue date Moreover the last annual MEF averages 175 days prior to the fiscal year-end post-

SOR and is not significantly closer to the fiscal year-end than the last annual MEF pre-SOR

With quarterly MEFs error and bias for the last forecast post-SOR are also significantly lower

than for the last forecast pre-SOR However quarterly MEF accuracy is similar for the first and

the last quarterly MEFs post-SOR Conversely accuracy appears to deteriorate between the first

and the last quarterly MEF pre-SOR For example pre-SOR error (bias) increases from 136

(051) for the first to 234 (106) for the last quarterly MEF The pre-SOR increase in error

(098) is significantly greater than the post-SOR decrease (-009) at the 10 level Horizon

is similar for the pre- and post-SOR last quarterly MEF While greater accuracy does not appear

to be attributed to shorter last MEF horizon we control for horizon in a multivariate setting35

Table 6 presents the multivariate results for MEF accuracy While we focus on accuracy

of last MEFs during the capital formation period in Columns 1 and 3 we also include last MEFs

that are not updated in the capital formation period in Columns 2 and 4 to determine if results are

similar We find SOR is associated with significantly lower error and bias for the last MEF

during the one month and one year prior to the SEO issue date (p = 003 and 003 for error p =

35 In untabulated results we perform the same analysis without restricting the last forecast to the prior month before

the SEO We find similar results for annual MEFs However we find no greater accuracy for quarterly MEFs when

they are not restricted to the prior one-month period

25

The multivariate results in Panel A of Table 4 are consistent with the univariate results

for occurrence frequency and 8-K size The significantly positive coefficient on SOR for the

forecasting sample in Columns 1ndash2 indicate that the likelihood of providing an annual MEF (p =

009) as well as guidance (p = 006) during the capital formation period is greater under SOR32

In addition Column 3 indicates guidance frequency in the capital formation period is greater

under SOR (p = 004) Overall the sign of the coefficients on the control variables loading

significantly are consistent with previous studies The significantly positive coefficients on SOR

(p = 001) in Columns 4ndash5 for the full sample show that both 8-Ks and total disclosures are more

frequent during the capital formation period under SOR The significantly positive coefficients

on SOR in Columns 6ndash7 for the full sample indicate that 8-K size of the last 8-K preceding the

SEO during the capital formation period is significantly larger under SOR even when

controlling for median size during the same period for WKSIs not issuing equity

[Insert Table 4 near here]

The multivariate results in Panel B provide additional evidence of shorter distance under

SOR for the last disclosure prior to the SEO issue date Columns 1ndash3 present results for last

disclosure restricted to the capital formation period The distance for last guidance is

significantly shorter (-563 days p = 003) and distance for last disclosure of any type is weakly

shorter (-106 days p = 020) under SOR The results in Columns 1ndash3 only partially capture the

relation between distance and SOR because the capital formation period restriction excludes the

last disclosure for a portion of the sample and does not account for the disproportionate

representation of the post-SOR sample Columns 4ndash6 remove the capital formation period

32 In untabulated results for the full sample the coefficient on SOR is not significant in a two-tailed test when

occurrence refers to an annual MEF (p = 017) but becomes significant when occurrence refers to an annual MEF or

FWP (p = 001) Thus greater guidance disclosure in the full sample during the capital formation period under SOR

stems from the use of FWPs and greater use of annual MEFs during the month prior to the SEO issue date

23

restriction and present results for last disclosure within one year prior to the issue date When

removing the restriction distance is significantly shorter for last 8-K (-691 days p = 002) and

last disclosure of any type (-372 days p = 004) under SOR Overall the results in Table 4

support the univariate results and indicate that SEO firms are providing more shorter distance

disclosure and thus more information during the month before the SEO issue date 33

53 Results for management earnings forecast accuracy

Table 5 presents univariate results for our second test Panel A presents error and bias

results for the last MEF (annual or annualized quarterly) within the one-month period prior to an

SEO to see if MEFs most likely to influence capital formation are more or less accurate post-

SOR34

The average forecast error is 70 lower under SOR (072 versus 243 p lt 001) In

addition average forecast bias is 83 smaller (133 versus 023 p = 009)

[Insert Table 5 near here]

To further explore the source of greater forecast accuracy Panel B (C) presents the

accuracy results for the first and the last annual (quarterly) MEF along with forecast horizons

Here first annual (quarterly) MEF is the first annual (quarterly) MEF provided during the one

year prior to the SEO issue date and last annual (quarterly) MEF is the last annual (quarterly)

MEF provided during the month before the SEO issue date We also present the difference in the

first and last annual (quarterly) MEF before and after SOR to see if firms are increasing or

decreasing their forecast accuracy as the SEO issue date approaches

33 In untabulated results we re-estimate all of our equations including additional measures that proxy for litigation

risk (Kim and Skinner 2012) to help ensure the SOR variable does not proxy for increased litigation risk Our

results for SOR remain robust when we include these additional measures Results are also robust to the inclusion of

SOR interacted with litigation risk The interaction is insignificant in the re-estimation of Table 4 34

We use annual MEF measures if a firm provides both annual and quarterly MEFs during the one-month period

prior to an SEO

24

The results indicate that greater accuracy for the last MEF under SOR is primarily driven

by increasing accuracy from first to last annual MEF and by maintaining as opposed to

decreasing accuracy from first to last quarterly MEF as the issue date approaches For example

error drops significantly (p = 001) from the first (136) to the last (077) annual MEF post-

SOR but not pre-SOR Likewise bias decreases significantly from 071 to 032 post-SOR

but not pre-SOR Thus while accuracy is somewhat greater for the first annual MEF under SOR

the significant improvement occurs with the last annual MEF within one month prior to the SEO

issue date Moreover the last annual MEF averages 175 days prior to the fiscal year-end post-

SOR and is not significantly closer to the fiscal year-end than the last annual MEF pre-SOR

With quarterly MEFs error and bias for the last forecast post-SOR are also significantly lower

than for the last forecast pre-SOR However quarterly MEF accuracy is similar for the first and

the last quarterly MEFs post-SOR Conversely accuracy appears to deteriorate between the first

and the last quarterly MEF pre-SOR For example pre-SOR error (bias) increases from 136

(051) for the first to 234 (106) for the last quarterly MEF The pre-SOR increase in error

(098) is significantly greater than the post-SOR decrease (-009) at the 10 level Horizon

is similar for the pre- and post-SOR last quarterly MEF While greater accuracy does not appear

to be attributed to shorter last MEF horizon we control for horizon in a multivariate setting35

Table 6 presents the multivariate results for MEF accuracy While we focus on accuracy

of last MEFs during the capital formation period in Columns 1 and 3 we also include last MEFs

that are not updated in the capital formation period in Columns 2 and 4 to determine if results are

similar We find SOR is associated with significantly lower error and bias for the last MEF

during the one month and one year prior to the SEO issue date (p = 003 and 003 for error p =

35 In untabulated results we perform the same analysis without restricting the last forecast to the prior month before

the SEO We find similar results for annual MEFs However we find no greater accuracy for quarterly MEFs when

they are not restricted to the prior one-month period

25

restriction and present results for last disclosure within one year prior to the issue date When

removing the restriction distance is significantly shorter for last 8-K (-691 days p = 002) and

last disclosure of any type (-372 days p = 004) under SOR Overall the results in Table 4

support the univariate results and indicate that SEO firms are providing more shorter distance

disclosure and thus more information during the month before the SEO issue date 33

53 Results for management earnings forecast accuracy

Table 5 presents univariate results for our second test Panel A presents error and bias

results for the last MEF (annual or annualized quarterly) within the one-month period prior to an

SEO to see if MEFs most likely to influence capital formation are more or less accurate post-

SOR34

The average forecast error is 70 lower under SOR (072 versus 243 p lt 001) In

addition average forecast bias is 83 smaller (133 versus 023 p = 009)

[Insert Table 5 near here]

To further explore the source of greater forecast accuracy Panel B (C) presents the

accuracy results for the first and the last annual (quarterly) MEF along with forecast horizons

Here first annual (quarterly) MEF is the first annual (quarterly) MEF provided during the one

year prior to the SEO issue date and last annual (quarterly) MEF is the last annual (quarterly)

MEF provided during the month before the SEO issue date We also present the difference in the

first and last annual (quarterly) MEF before and after SOR to see if firms are increasing or

decreasing their forecast accuracy as the SEO issue date approaches

33 In untabulated results we re-estimate all of our equations including additional measures that proxy for litigation

risk (Kim and Skinner 2012) to help ensure the SOR variable does not proxy for increased litigation risk Our

results for SOR remain robust when we include these additional measures Results are also robust to the inclusion of

SOR interacted with litigation risk The interaction is insignificant in the re-estimation of Table 4 34

We use annual MEF measures if a firm provides both annual and quarterly MEFs during the one-month period

prior to an SEO

24

The results indicate that greater accuracy for the last MEF under SOR is primarily driven

by increasing accuracy from first to last annual MEF and by maintaining as opposed to

decreasing accuracy from first to last quarterly MEF as the issue date approaches For example

error drops significantly (p = 001) from the first (136) to the last (077) annual MEF post-

SOR but not pre-SOR Likewise bias decreases significantly from 071 to 032 post-SOR

but not pre-SOR Thus while accuracy is somewhat greater for the first annual MEF under SOR

the significant improvement occurs with the last annual MEF within one month prior to the SEO

issue date Moreover the last annual MEF averages 175 days prior to the fiscal year-end post-

SOR and is not significantly closer to the fiscal year-end than the last annual MEF pre-SOR

With quarterly MEFs error and bias for the last forecast post-SOR are also significantly lower

than for the last forecast pre-SOR However quarterly MEF accuracy is similar for the first and

the last quarterly MEFs post-SOR Conversely accuracy appears to deteriorate between the first

and the last quarterly MEF pre-SOR For example pre-SOR error (bias) increases from 136

(051) for the first to 234 (106) for the last quarterly MEF The pre-SOR increase in error

(098) is significantly greater than the post-SOR decrease (-009) at the 10 level Horizon

is similar for the pre- and post-SOR last quarterly MEF While greater accuracy does not appear

to be attributed to shorter last MEF horizon we control for horizon in a multivariate setting35

Table 6 presents the multivariate results for MEF accuracy While we focus on accuracy

of last MEFs during the capital formation period in Columns 1 and 3 we also include last MEFs

that are not updated in the capital formation period in Columns 2 and 4 to determine if results are

similar We find SOR is associated with significantly lower error and bias for the last MEF

during the one month and one year prior to the SEO issue date (p = 003 and 003 for error p =

35 In untabulated results we perform the same analysis without restricting the last forecast to the prior month before

the SEO We find similar results for annual MEFs However we find no greater accuracy for quarterly MEFs when

they are not restricted to the prior one-month period

25

The results indicate that greater accuracy for the last MEF under SOR is primarily driven

by increasing accuracy from first to last annual MEF and by maintaining as opposed to

decreasing accuracy from first to last quarterly MEF as the issue date approaches For example

error drops significantly (p = 001) from the first (136) to the last (077) annual MEF post-

SOR but not pre-SOR Likewise bias decreases significantly from 071 to 032 post-SOR

but not pre-SOR Thus while accuracy is somewhat greater for the first annual MEF under SOR

the significant improvement occurs with the last annual MEF within one month prior to the SEO

issue date Moreover the last annual MEF averages 175 days prior to the fiscal year-end post-

SOR and is not significantly closer to the fiscal year-end than the last annual MEF pre-SOR

With quarterly MEFs error and bias for the last forecast post-SOR are also significantly lower

than for the last forecast pre-SOR However quarterly MEF accuracy is similar for the first and

the last quarterly MEFs post-SOR Conversely accuracy appears to deteriorate between the first

and the last quarterly MEF pre-SOR For example pre-SOR error (bias) increases from 136

(051) for the first to 234 (106) for the last quarterly MEF The pre-SOR increase in error

(098) is significantly greater than the post-SOR decrease (-009) at the 10 level Horizon

is similar for the pre- and post-SOR last quarterly MEF While greater accuracy does not appear

to be attributed to shorter last MEF horizon we control for horizon in a multivariate setting35

Table 6 presents the multivariate results for MEF accuracy While we focus on accuracy

of last MEFs during the capital formation period in Columns 1 and 3 we also include last MEFs

that are not updated in the capital formation period in Columns 2 and 4 to determine if results are

similar We find SOR is associated with significantly lower error and bias for the last MEF

during the one month and one year prior to the SEO issue date (p = 003 and 003 for error p =

35 In untabulated results we perform the same analysis without restricting the last forecast to the prior month before

the SEO We find similar results for annual MEFs However we find no greater accuracy for quarterly MEFs when

they are not restricted to the prior one-month period

25

006 and 001 for bias)36

Overall we find no evidence that managers increase MEF optimism

under SOR In fact the results suggest that managers provide less optimistic and more accurate

MEFs before an SEO post-SOR despite the temptation of temporarily influencing stock prices

through rosy forecasts In untabulated results we find that MEFs provided during the month

before the issue date post-SOR have a greater proportion of downward revisions which gives

additional evidence against opportunistic MEFs Falling profits during the 2008 financial crisis

do not appear to explain the downward revisions Less than 6 of the post-SOR subsample with

downward revisions ie SEOs with last MEFs lower than first MEFs have issue dates in 2008

Further we find that post-SOR downward revisers tend to raise larger SEO proceeds and have

greater firm risk both of which are associated with greater litigation risk These results suggest

instead that firms may be correcting expectations downward prior to an SEO to mitigate

litigation risk37

Although managers may delay releasing bad news on average (Kothari et al

2009) fear of litigation may prompt revising MEFs downward prior to an SEO (Skinner 1994)

[Insert Table 6 near here]

54 Results for information magnitude and stock returns

For our third test Fig 2 plots the cumulative magnitude (ie absolute market-adjusted

returns) beginning 20 trading days before and ending on the SEO issue date which approximates

the capital formation period for the pre- and post-SOR periods Consistent with greater

information reflected in stock prices as the issue date approaches under SOR the cumulative

36 In untabulated results we interact SOR with litigation risk All results are robust and the interaction is significant

only when estimating bias for MEFs provided during the prior month The coefficients on SOR and litigation risk

remain negative and significant while the interaction is significantly positive indicating SOR and litigation risk are

both associated with lower bias but the lower bias after SOR is outside litigation risk industries 37

Previous studies find increased earnings management prior to SEOs (eg Cohen and Zarowin 2010) We do not

believe earnings management is responsible for greater accuracy under SOR for two reasons First street earnings

provide more accurate operating earnings for valuation Second we find a greater incidence of downward revisions

instead of upward revisions prior to the SEO

26

magnitude post-SOR plot lies above the pre-SOR plot and the gap widens as the issue date

approaches The univariate results in Table 7 Panel A indicate that the post-SOR magnitude [-

200] of 5520 is 47 greater at issue than the pre-SOR value of 3756 (p lt 001) Moreover

the post-SOR abnormal magnitude of 1602 is 87 greater than the pre-SOR value of 855

indicating that the magnitude of information during the capital formation period is significantly

greater under SOR even when adjusting for factors other than capital formation disclosure that

may influence magnitude during the period surrounding an SEO

[Insert Fig 2 and Table 7 near here]

Panel B compares magnitude for the pre-SOR full sample with post-SOR subsamples

based on whether the SEO is preceded by any type of disclosure or an 8-K within one week

defined as five days to see if the increasing magnitude of information as the issue date

approaches post-SOR in Fig 2 is related to disclosure immediately preceding the issue date

Taken together the results for magnitude [-200] and abnormal magnitude provide additional

support that greater information (ie absolute market-adjusted returns) during the capital

formation period under SOR is associated with disclosure Magnitude [-200] is significantly

greater post-SOR regardless of whether SEOs have short distance disclosure (ie disclosure

within one week before the issue date) However abnormal magnitude is significantly greater

only when SEOs have short distance disclosure Average abnormal magnitude is 2008

(2102) when an SEO is immediately preceded by any form of disclosure (an 8-K) post-SOR

compared with 855 for the pre-SOR sample38

38 We analyze the content distribution of the various items reported on the last 8 -K filed within a week prior to the

issue date (short distance 8-K) post-SOR to provide a contextual analysis of these communications (untabulated)

We find over 80 of these 8-Ks include Financial Statements and Exhibits (Item 901) and about 20 include one

or more of the following Regulation FD Disclosure (Item 701) Results of Operations and Financial Condition

(Item 202) and Entry into a Material Definitive Agreement (Item 101) the last of which includes underwriting

agreements related to the SEO More than half include Other Events (Item 801) The SEC encourages firms to

disclose information on Item 801 that is of importance to security holders but disclosure of these items is not

27

Fig 3 provides additional evidence that greater magnitude post-SOR is related to 8-Ks

during the capital formation period The figure plots three-day [-1+1] magnitude aggregated for

all 8-K filing dates during the capital formation period pre- and post-SOR Similar to Fig 2 it

shows that 8-Ks are associated with greater information magnitude being reflected in stock

prices post-SOR with the difference increasing as the SEO issue date approaches

[Insert Fig 3 near here]

Results for return are presented in Table 839

Overall abnormal return (Panel A) is more

than five times greater under SOR (472 versus 085) The difference is significantly greater

in a one-tailed (two-tailed) test for the full (non-forecasting) sample Greater abnormal return

results from greater return [-200] 503 versus 245 combined with a lower positive drift

after the issue return [+1+20] of 031 versus 160 Similar to the results for magnitude

greater abnormal return is driven by SEOs preceded by any form of disclosure or 8-Ks within

one week Abnormal return (Panel B) is significantly greater in a two-tailed test for the post-

SOR subsamples with any disclosure or an 8-K immediately preceding the SEO (748 and

793 versus 085 pre-SOR) while the differences are insignificant for the post-SOR

subsamples with no disclosure or no 8-K within one week (188 and 188 versus 085 pre-

SOR) Furthermore greater abnormal return in the post-SOR subsamples is primarily driven by

greater positive stock returns during the capital formation period (return [-200] of 836 and

936 versus 245 pre-SOR) No evidence exists of reversal after the issue date in these

subsamples (return [+1+20] is 088 and 143) Higher stock returns during the capital

mandatory (Lerman and Livnat 2010 Segal and Segal 2013) We find that Other Events on these 8-Ks include

information such as press releases MEFs and other forward-looking information These results suggest firms are

providing a variety of information that appears to be important to equity investors during the capital formation

period under SOR 39

We use raw returns because we are interested in the net effects for the firms However we find similar results

when we use cumulative market-adjusted returns In untabulated results we also control for size and book-to-market

in a multivariate setting and continue to find significant results for SOR interacted with short distance disclosure

28

formation period with no reversal afterward is consistent with greater capital formation benefits

through more informative disclosure and a richer information environment under SOR

[Insert Table 8 near here]

6 Conclusion

This paper is the first to examine differences in disclosure and the information

environment at the time of seasoned equity capital formation under the Securities Offering

Reform which relaxes disclosure restrictions prior to an SEO We examine whether SOR is

associated with more factual and forward-looking disclosure by examining 8-K file size and the

occurrence frequency and timing of multiple forms of disclosure for a sample of WKSIs who

are the most economically significant issuers and receive the greatest benefits under SOR We

also investigate whether SOR is associated with a more informative or opportunistic information

environment by analyzing management earnings forecast accuracy as well as cumulative

absolute market-adjusted returns and cumulative signed raw returns

Our research design differs from other studies in important ways First we use the issue

date rather than the filing date as the SEO date in response to changes in the registration process

under SOR Second we examine multiple forms of disclosure during the one-month period prior

to the SEO issue date ie the capital formation period In addition to MEFs and earnings

releases we analyze 8-Ks and a new form of disclosure introduced with SOR free writing

prospectuses Finally our sample includes non-forecasting firms in addition to forecasting firms

which allows us to provide insight into capital formation period disclosure by both firms with

and firms without a history of forecasting

29

We find greater forward-looking and factual disclosure during the equity capital

formation period under SOR However despite the predictions by critics of SOR we find no

evidence that relaxed disclosure restrictions under SOR are associated with opportunistic

disclosure Within the post-SOR forecasting sample MEFs are more accurate and greater

disclosure and closer proximity are strongest for guidance disclosure In contrast greater

disclosure and closer proximity in the post-SOR non-forecasting sample are primarily driven by

8-K disclosure though some non-forecasting firms do provide guidance disclosure through the

newly permitted FWPs We also find that greater absolute market-adjusted returns under SOR

are highest in the non-forecasting sample and for SEOs with an 8-K filed within a week of the

issue date Moreover significantly more positive stock returns are evident during the capital

formation period with no reversal afterward for SEOs with an 8-K filed within a week of the

issue date This is of particular interest given that the majority of SEO firms continue to provide

no MEFs under SOR One caveat to our results is that the power of certain tests may be limited

by sample size during our sample period As the period since SOR lengthens two fruitful areas

for future research would be to investigate whether disclosure under SOR changes over time and

whether SOR affects a firmrsquos ability to raise capital in different economic cycles

Overall the results in this paper suggest that relaxed disclosure restrictions and reduced

regulatory uncertainty under SOR are associated with greater disclosure immediately preceding

the SEO issue date when investors assess the SEO price and decide whether to invest

Disclosure during this time especially 8-K disclosure is related to a richer information

environment with capital formation benefits Thus our paper contributes to the literature on

disclosure SEOs and regulation

30

Appendix Variable definitions and other key terms

Variable Definition

Abnormal magnitude

Abnormal return

Analyst dispersion

Analyst following

Annual MEF

Bias

Book-to-market

Closest disclosure

Distance

Earnings announcement

Error

Firm risk

First annual MEF

First quarterly MEF

Frequency

Forecasting

FWP

Guidance

Horizon

Institutional ownership

Last annual MEF

Last MEF

The magnitude over trading days [-200] less the magnitude over trading days [+1+20] The return over trading days [-200] less the return over trading days [+1+20] The standard deviation of analystsrsquo annual earnings estimates divided by the absolute value of the median analyst earnings forecast Analyst dispersion is calculated using the last reported analyst estimates in Institutional Brokersrsquo Estimate System (IBES) for the month-end at least 30 days prior to the seasoned equity offering (SEO) issue date The number of analysts providing annual forecasts at the month ending at least 30 days prior to the SEO issue date A firm-issued point or median of the range forecast of annual earnings The firmrsquos annual MEF or quarterly MEF minus the actual earnings divided by the stock price at the beginning of the fiscal period annualized The book value of common equity divided by the market value of common equity at the fiscal year-end prior to the SEO issue date Book-to-market is set to the minimum sample value for two observations with missing values The number of days between the last disclosure of the following type prior to the SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement The number of days between the specified disclosure and the SEO issue date The periodic announcement of quarterly or annual earnings during the 30 days prior to the SEO issue date The absolute value of the firmrsquos annual MEF or quarterly MEF minus the actual earnings divided by the stock price at the beginning of the fiscal period annualized The standard deviation of the firms daily returns over the period [-250-20] trading days to the SEO issue date The annual MEF issued farthest from the SEO within 365 days of the SEO issue date The quarterly MEF issued farthest from the SEO within 365 days of the SEO issue date The number of the specified disclosures during the month prior to the SEO issue date An indicator variable equal to one if the firm provides an annual MEF or quarterly MEF during the 365 days prior to the SEO issue date A free writing prospectus is an SEC filing of communication FWPs can include any written communication in printed form and other communication via television radio broadcasts and certain electronic road shows (Pena 2007) A forward-looking disclosure in the form of an MEF or FWP The number of days between the annual MEF or quarterly MEF and the fiscal period end annualized The percentage of the firmrsquos outstanding shares held by institutions at the quarter-end prior to the SEO issue date The annual MEF issued closest to the SEO within 30 days of the SEO issue date The last quarterly MEF or annual MEF within 30 days of the SEO issue date

31

Last quarterly MEF The quarterly MEF issued closest to the SEO within 30 days of the SEO issue date

Litigation risk An indicator variable equal to one if the SEO firm operates in the following high litigation risk industries (standard

industrial classification codes) biotechnology (2833ndash2836) research and development services (8731ndash8734)

programming (7371ndash7379) computers (3570ndash3577) electronics (3600ndash3674) or retailing (5200ndash5961)

Ln(market capitalization) The natural log of the market value of common equity at fiscal year-end prior to the SEO issue date

Loss An indicator variable equal to one if the firm reports a loss in the current fiscal year

Magnitude The absolute value of the daily return net of the return on the CRSP value-weighted index cumulated over the

specified trading days [-200] and [+0+20] with day 0 being the volume-based corrected issue date (Corwin 2003)

Market capitalization The market value of common equity at the fiscal year-end prior to the SEO issue date in billions of dollars

MEF A management earnings forecast a firm-issued point or median of range forecast of annual or quarterly earnings

Non-forecasting An indicator variable equal to one if the firm does not provide an annual MEF or quarterly MEF during the 365 days

Non-SEO 8-K size

prior to the SEO issue date

The median 8-K size for non-SEO well-known seasoned issuers (WKSIs) during the same period Non-SEO WKSIs

meet the WKSI market capitalization criteria but do not conduct an SEO during the same calendar year

Occurrence An indicator variable equal to one if the firm issues the specified voluntary disclosure (annual MEF quarterly MEF

or FWP) within 30 days prior to the SEO issue date

Pre-issue days The number of days between the filing of a preliminary prospectus (pre-issue date) and the SEO issue date

Quarterly MEF A firm-issued point or median of range forecast of quarterly earnings

Registration days The number of days between the filing of a registration statement and the SEO issue date

Residual analyst following The residual value after regressing analyst following on Ln(market capitalization)

Return

SEO proceeds

SEO size

Shelf

Short distance 8-K

Short distance disclosure

SOR

Total disclosure

The raw daily return cumulated over the specified trading days [-200] and [+0+20] with day 0 being the volume-

based corrected issue date (Corwin 2003)

The gross proceeds from the seasoned equity offering before underwriting

The SEO proceeds scaled by market capitalization at the fiscal year-end prior to the SEO issue date

An indicator variable equal to one if the SEO is offered from a shelf registration

An indicator variable equal to one if the firm provides an 8-K during the five days prior to the SEO issue date

An indicator variable equal to one if the firm provides any of the following disclosures during the five days prior to the

SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement

An indicator variable equal to one if the SEO issue date is on or after December 1 2005 when SOR became effective

The aggregate number of annual MEFs quarterly MEFs FWPs 8-Ks and earnings announcements within 30 days of

the SEO issue date

8-K abnormal size The SEO firmrsquos 8-K size less the median non-SEO 8-K size

8-K size The size in megabytes of the last 8-K during the 30 days prior to the SEO issue date otherwise set to zero

32

References

Aboody D Kasznik R 2000 CEO stock option awards and the timing of voluntary corporate

disclosures J of Account and Econ 29 (1) 73ndash100

Ajinkya B Bhojraj S Sengupta P 2005 The association between outside directors institutional

investors and the properties of management earnings forecasts J of Account Res 43 (3) 343ndash 376

Aldrich JH Nelson FD 1984 Linear Probability Logit and Probit Models SAGE Publications

Beverly Hills CA

Ali A Klasa S Yeung E 2014 Industry concentration and corporate disclosure policy Working

paper The University of Texas at Dallas

Asquith P Mullins DW 1986 Equity issues and offering dilution J of Financ Econ 15 (1) 61ndash89

Autore DM Kumar R Shome DK 2008 The revival of shelf-registered corporate equity offerings

J of Corp Financ 14 (1) 32ndash50

Bailey W Karolyi GA Salva C 2006 The economic consequences of increased disclosure Evidence

from international cross-listings J of Financ Econ 81 (1) 175ndash213

Beck T Levine R Levkov A 2010 Big bad banks The winners and losers from bank deregulation in

the United States J of Financ 65 (5) 1637ndash1667

Bethel JE Krigman L 2008 Managing the costs of issuing common equity The role of registration

choice Q J of Financ and Account 47 (4) 57ndash85

Beyer A Cohen DA Lys TZ Walther BR 2010 The financial reporting environment Review of

the recent literature J of Account and Econ 50 (2ndash3) 179ndash234

Blackwell DW Marr MW Spivey MF 1990 Shelf registration and the reduced due diligence

argument Implications of the underwriter certification and the implicit insurance hypotheses J

of Financ and Quant Anal 25 (2) 245ndash259

Botosan CA 1997 Disclosure level and the cost of equity capital Account Rev 72 (3) 323ndash349

Brockman P Khurana IK Martin X 2008 Voluntary disclosures around share repurchases J of

Financ Econ 89 (1) 175ndash191

Cadwalader Wickersham amp Taft LLP 2005 New SEC rules for Securities Offering Reform August 4

httpwwwcadwaladercomassetsclient_friend080405SECRulesforSecuritiesOffering

Reformpdf [Accessed 16 Dec 2011]

Chang X Dasgupta S Hillary G 2006 Analyst coverage and financing decisions J of Financ 61 (6)

2009ndash2048

Cheng Q Lo K 2006 Insider trading and voluntary disclosures J of Account Res 44 (5) 815ndash848

33

formation period with no reversal afterward is consistent with greater capital formation benefits

through more informative disclosure and a richer information environment under SOR

[Insert Table 8 near here]

6 Conclusion

This paper is the first to examine differences in disclosure and the information

environment at the time of seasoned equity capital formation under the Securities Offering

Reform which relaxes disclosure restrictions prior to an SEO We examine whether SOR is

associated with more factual and forward-looking disclosure by examining 8-K file size and the

occurrence frequency and timing of multiple forms of disclosure for a sample of WKSIs who

are the most economically significant issuers and receive the greatest benefits under SOR We

also investigate whether SOR is associated with a more informative or opportunistic information

environment by analyzing management earnings forecast accuracy as well as cumulative

absolute market-adjusted returns and cumulative signed raw returns

Our research design differs from other studies in important ways First we use the issue

date rather than the filing date as the SEO date in response to changes in the registration process

under SOR Second we examine multiple forms of disclosure during the one-month period prior

to the SEO issue date ie the capital formation period In addition to MEFs and earnings

releases we analyze 8-Ks and a new form of disclosure introduced with SOR free writing

prospectuses Finally our sample includes non-forecasting firms in addition to forecasting firms

which allows us to provide insight into capital formation period disclosure by both firms with

and firms without a history of forecasting

29

We find greater forward-looking and factual disclosure during the equity capital

formation period under SOR However despite the predictions by critics of SOR we find no

evidence that relaxed disclosure restrictions under SOR are associated with opportunistic

disclosure Within the post-SOR forecasting sample MEFs are more accurate and greater

disclosure and closer proximity are strongest for guidance disclosure In contrast greater

disclosure and closer proximity in the post-SOR non-forecasting sample are primarily driven by

8-K disclosure though some non-forecasting firms do provide guidance disclosure through the

newly permitted FWPs We also find that greater absolute market-adjusted returns under SOR

are highest in the non-forecasting sample and for SEOs with an 8-K filed within a week of the

issue date Moreover significantly more positive stock returns are evident during the capital

formation period with no reversal afterward for SEOs with an 8-K filed within a week of the

issue date This is of particular interest given that the majority of SEO firms continue to provide

no MEFs under SOR One caveat to our results is that the power of certain tests may be limited

by sample size during our sample period As the period since SOR lengthens two fruitful areas

for future research would be to investigate whether disclosure under SOR changes over time and

whether SOR affects a firmrsquos ability to raise capital in different economic cycles

Overall the results in this paper suggest that relaxed disclosure restrictions and reduced

regulatory uncertainty under SOR are associated with greater disclosure immediately preceding

the SEO issue date when investors assess the SEO price and decide whether to invest

Disclosure during this time especially 8-K disclosure is related to a richer information

environment with capital formation benefits Thus our paper contributes to the literature on

disclosure SEOs and regulation

30

Appendix Variable definitions and other key terms

Variable Definition

Abnormal magnitude

Abnormal return

Analyst dispersion

Analyst following

Annual MEF

Bias

Book-to-market

Closest disclosure

Distance

Earnings announcement

Error

Firm risk

First annual MEF

First quarterly MEF

Frequency

Forecasting

FWP

Guidance

Horizon

Institutional ownership

Last annual MEF

Last MEF

The magnitude over trading days [-200] less the magnitude over trading days [+1+20] The return over trading days [-200] less the return over trading days [+1+20] The standard deviation of analystsrsquo annual earnings estimates divided by the absolute value of the median analyst earnings forecast Analyst dispersion is calculated using the last reported analyst estimates in Institutional Brokersrsquo Estimate System (IBES) for the month-end at least 30 days prior to the seasoned equity offering (SEO) issue date The number of analysts providing annual forecasts at the month ending at least 30 days prior to the SEO issue date A firm-issued point or median of the range forecast of annual earnings The firmrsquos annual MEF or quarterly MEF minus the actual earnings divided by the stock price at the beginning of the fiscal period annualized The book value of common equity divided by the market value of common equity at the fiscal year-end prior to the SEO issue date Book-to-market is set to the minimum sample value for two observations with missing values The number of days between the last disclosure of the following type prior to the SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement The number of days between the specified disclosure and the SEO issue date The periodic announcement of quarterly or annual earnings during the 30 days prior to the SEO issue date The absolute value of the firmrsquos annual MEF or quarterly MEF minus the actual earnings divided by the stock price at the beginning of the fiscal period annualized The standard deviation of the firms daily returns over the period [-250-20] trading days to the SEO issue date The annual MEF issued farthest from the SEO within 365 days of the SEO issue date The quarterly MEF issued farthest from the SEO within 365 days of the SEO issue date The number of the specified disclosures during the month prior to the SEO issue date An indicator variable equal to one if the firm provides an annual MEF or quarterly MEF during the 365 days prior to the SEO issue date A free writing prospectus is an SEC filing of communication FWPs can include any written communication in printed form and other communication via television radio broadcasts and certain electronic road shows (Pena 2007) A forward-looking disclosure in the form of an MEF or FWP The number of days between the annual MEF or quarterly MEF and the fiscal period end annualized The percentage of the firmrsquos outstanding shares held by institutions at the quarter-end prior to the SEO issue date The annual MEF issued closest to the SEO within 30 days of the SEO issue date The last quarterly MEF or annual MEF within 30 days of the SEO issue date

31

Last quarterly MEF The quarterly MEF issued closest to the SEO within 30 days of the SEO issue date

Litigation risk An indicator variable equal to one if the SEO firm operates in the following high litigation risk industries (standard

industrial classification codes) biotechnology (2833ndash2836) research and development services (8731ndash8734)

programming (7371ndash7379) computers (3570ndash3577) electronics (3600ndash3674) or retailing (5200ndash5961)

Ln(market capitalization) The natural log of the market value of common equity at fiscal year-end prior to the SEO issue date

Loss An indicator variable equal to one if the firm reports a loss in the current fiscal year

Magnitude The absolute value of the daily return net of the return on the CRSP value-weighted index cumulated over the

specified trading days [-200] and [+0+20] with day 0 being the volume-based corrected issue date (Corwin 2003)

Market capitalization The market value of common equity at the fiscal year-end prior to the SEO issue date in billions of dollars

MEF A management earnings forecast a firm-issued point or median of range forecast of annual or quarterly earnings

Non-forecasting An indicator variable equal to one if the firm does not provide an annual MEF or quarterly MEF during the 365 days

Non-SEO 8-K size

prior to the SEO issue date

The median 8-K size for non-SEO well-known seasoned issuers (WKSIs) during the same period Non-SEO WKSIs

meet the WKSI market capitalization criteria but do not conduct an SEO during the same calendar year

Occurrence An indicator variable equal to one if the firm issues the specified voluntary disclosure (annual MEF quarterly MEF

or FWP) within 30 days prior to the SEO issue date

Pre-issue days The number of days between the filing of a preliminary prospectus (pre-issue date) and the SEO issue date

Quarterly MEF A firm-issued point or median of range forecast of quarterly earnings

Registration days The number of days between the filing of a registration statement and the SEO issue date

Residual analyst following The residual value after regressing analyst following on Ln(market capitalization)

Return

SEO proceeds

SEO size

Shelf

Short distance 8-K

Short distance disclosure

SOR

Total disclosure

The raw daily return cumulated over the specified trading days [-200] and [+0+20] with day 0 being the volume-

based corrected issue date (Corwin 2003)

The gross proceeds from the seasoned equity offering before underwriting

The SEO proceeds scaled by market capitalization at the fiscal year-end prior to the SEO issue date

An indicator variable equal to one if the SEO is offered from a shelf registration

An indicator variable equal to one if the firm provides an 8-K during the five days prior to the SEO issue date

An indicator variable equal to one if the firm provides any of the following disclosures during the five days prior to the

SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement

An indicator variable equal to one if the SEO issue date is on or after December 1 2005 when SOR became effective

The aggregate number of annual MEFs quarterly MEFs FWPs 8-Ks and earnings announcements within 30 days of

the SEO issue date

8-K abnormal size The SEO firmrsquos 8-K size less the median non-SEO 8-K size

8-K size The size in megabytes of the last 8-K during the 30 days prior to the SEO issue date otherwise set to zero

32

References

Aboody D Kasznik R 2000 CEO stock option awards and the timing of voluntary corporate

disclosures J of Account and Econ 29 (1) 73ndash100

Ajinkya B Bhojraj S Sengupta P 2005 The association between outside directors institutional

investors and the properties of management earnings forecasts J of Account Res 43 (3) 343ndash 376

Aldrich JH Nelson FD 1984 Linear Probability Logit and Probit Models SAGE Publications

Beverly Hills CA

Ali A Klasa S Yeung E 2014 Industry concentration and corporate disclosure policy Working

paper The University of Texas at Dallas

Asquith P Mullins DW 1986 Equity issues and offering dilution J of Financ Econ 15 (1) 61ndash89

Autore DM Kumar R Shome DK 2008 The revival of shelf-registered corporate equity offerings

J of Corp Financ 14 (1) 32ndash50

Bailey W Karolyi GA Salva C 2006 The economic consequences of increased disclosure Evidence

from international cross-listings J of Financ Econ 81 (1) 175ndash213

Beck T Levine R Levkov A 2010 Big bad banks The winners and losers from bank deregulation in

the United States J of Financ 65 (5) 1637ndash1667

Bethel JE Krigman L 2008 Managing the costs of issuing common equity The role of registration

choice Q J of Financ and Account 47 (4) 57ndash85

Beyer A Cohen DA Lys TZ Walther BR 2010 The financial reporting environment Review of

the recent literature J of Account and Econ 50 (2ndash3) 179ndash234

Blackwell DW Marr MW Spivey MF 1990 Shelf registration and the reduced due diligence

argument Implications of the underwriter certification and the implicit insurance hypotheses J

of Financ and Quant Anal 25 (2) 245ndash259

Botosan CA 1997 Disclosure level and the cost of equity capital Account Rev 72 (3) 323ndash349

Brockman P Khurana IK Martin X 2008 Voluntary disclosures around share repurchases J of

Financ Econ 89 (1) 175ndash191

Cadwalader Wickersham amp Taft LLP 2005 New SEC rules for Securities Offering Reform August 4

httpwwwcadwaladercomassetsclient_friend080405SECRulesforSecuritiesOffering

Reformpdf [Accessed 16 Dec 2011]

Chang X Dasgupta S Hillary G 2006 Analyst coverage and financing decisions J of Financ 61 (6)

2009ndash2048

Cheng Q Lo K 2006 Insider trading and voluntary disclosures J of Account Res 44 (5) 815ndash848

33

We find greater forward-looking and factual disclosure during the equity capital

formation period under SOR However despite the predictions by critics of SOR we find no

evidence that relaxed disclosure restrictions under SOR are associated with opportunistic

disclosure Within the post-SOR forecasting sample MEFs are more accurate and greater

disclosure and closer proximity are strongest for guidance disclosure In contrast greater

disclosure and closer proximity in the post-SOR non-forecasting sample are primarily driven by

8-K disclosure though some non-forecasting firms do provide guidance disclosure through the

newly permitted FWPs We also find that greater absolute market-adjusted returns under SOR

are highest in the non-forecasting sample and for SEOs with an 8-K filed within a week of the

issue date Moreover significantly more positive stock returns are evident during the capital

formation period with no reversal afterward for SEOs with an 8-K filed within a week of the

issue date This is of particular interest given that the majority of SEO firms continue to provide

no MEFs under SOR One caveat to our results is that the power of certain tests may be limited

by sample size during our sample period As the period since SOR lengthens two fruitful areas

for future research would be to investigate whether disclosure under SOR changes over time and

whether SOR affects a firmrsquos ability to raise capital in different economic cycles

Overall the results in this paper suggest that relaxed disclosure restrictions and reduced

regulatory uncertainty under SOR are associated with greater disclosure immediately preceding

the SEO issue date when investors assess the SEO price and decide whether to invest

Disclosure during this time especially 8-K disclosure is related to a richer information

environment with capital formation benefits Thus our paper contributes to the literature on

disclosure SEOs and regulation

30

Appendix Variable definitions and other key terms

Variable Definition

Abnormal magnitude

Abnormal return

Analyst dispersion

Analyst following

Annual MEF

Bias

Book-to-market

Closest disclosure

Distance

Earnings announcement

Error

Firm risk

First annual MEF

First quarterly MEF

Frequency

Forecasting

FWP

Guidance

Horizon

Institutional ownership

Last annual MEF

Last MEF

The magnitude over trading days [-200] less the magnitude over trading days [+1+20] The return over trading days [-200] less the return over trading days [+1+20] The standard deviation of analystsrsquo annual earnings estimates divided by the absolute value of the median analyst earnings forecast Analyst dispersion is calculated using the last reported analyst estimates in Institutional Brokersrsquo Estimate System (IBES) for the month-end at least 30 days prior to the seasoned equity offering (SEO) issue date The number of analysts providing annual forecasts at the month ending at least 30 days prior to the SEO issue date A firm-issued point or median of the range forecast of annual earnings The firmrsquos annual MEF or quarterly MEF minus the actual earnings divided by the stock price at the beginning of the fiscal period annualized The book value of common equity divided by the market value of common equity at the fiscal year-end prior to the SEO issue date Book-to-market is set to the minimum sample value for two observations with missing values The number of days between the last disclosure of the following type prior to the SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement The number of days between the specified disclosure and the SEO issue date The periodic announcement of quarterly or annual earnings during the 30 days prior to the SEO issue date The absolute value of the firmrsquos annual MEF or quarterly MEF minus the actual earnings divided by the stock price at the beginning of the fiscal period annualized The standard deviation of the firms daily returns over the period [-250-20] trading days to the SEO issue date The annual MEF issued farthest from the SEO within 365 days of the SEO issue date The quarterly MEF issued farthest from the SEO within 365 days of the SEO issue date The number of the specified disclosures during the month prior to the SEO issue date An indicator variable equal to one if the firm provides an annual MEF or quarterly MEF during the 365 days prior to the SEO issue date A free writing prospectus is an SEC filing of communication FWPs can include any written communication in printed form and other communication via television radio broadcasts and certain electronic road shows (Pena 2007) A forward-looking disclosure in the form of an MEF or FWP The number of days between the annual MEF or quarterly MEF and the fiscal period end annualized The percentage of the firmrsquos outstanding shares held by institutions at the quarter-end prior to the SEO issue date The annual MEF issued closest to the SEO within 30 days of the SEO issue date The last quarterly MEF or annual MEF within 30 days of the SEO issue date

31

Last quarterly MEF The quarterly MEF issued closest to the SEO within 30 days of the SEO issue date

Litigation risk An indicator variable equal to one if the SEO firm operates in the following high litigation risk industries (standard

industrial classification codes) biotechnology (2833ndash2836) research and development services (8731ndash8734)

programming (7371ndash7379) computers (3570ndash3577) electronics (3600ndash3674) or retailing (5200ndash5961)

Ln(market capitalization) The natural log of the market value of common equity at fiscal year-end prior to the SEO issue date

Loss An indicator variable equal to one if the firm reports a loss in the current fiscal year

Magnitude The absolute value of the daily return net of the return on the CRSP value-weighted index cumulated over the

specified trading days [-200] and [+0+20] with day 0 being the volume-based corrected issue date (Corwin 2003)

Market capitalization The market value of common equity at the fiscal year-end prior to the SEO issue date in billions of dollars

MEF A management earnings forecast a firm-issued point or median of range forecast of annual or quarterly earnings

Non-forecasting An indicator variable equal to one if the firm does not provide an annual MEF or quarterly MEF during the 365 days

Non-SEO 8-K size

prior to the SEO issue date

The median 8-K size for non-SEO well-known seasoned issuers (WKSIs) during the same period Non-SEO WKSIs

meet the WKSI market capitalization criteria but do not conduct an SEO during the same calendar year

Occurrence An indicator variable equal to one if the firm issues the specified voluntary disclosure (annual MEF quarterly MEF

or FWP) within 30 days prior to the SEO issue date

Pre-issue days The number of days between the filing of a preliminary prospectus (pre-issue date) and the SEO issue date

Quarterly MEF A firm-issued point or median of range forecast of quarterly earnings

Registration days The number of days between the filing of a registration statement and the SEO issue date

Residual analyst following The residual value after regressing analyst following on Ln(market capitalization)

Return

SEO proceeds

SEO size

Shelf

Short distance 8-K

Short distance disclosure

SOR

Total disclosure

The raw daily return cumulated over the specified trading days [-200] and [+0+20] with day 0 being the volume-

based corrected issue date (Corwin 2003)

The gross proceeds from the seasoned equity offering before underwriting

The SEO proceeds scaled by market capitalization at the fiscal year-end prior to the SEO issue date

An indicator variable equal to one if the SEO is offered from a shelf registration

An indicator variable equal to one if the firm provides an 8-K during the five days prior to the SEO issue date

An indicator variable equal to one if the firm provides any of the following disclosures during the five days prior to the

SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement

An indicator variable equal to one if the SEO issue date is on or after December 1 2005 when SOR became effective

The aggregate number of annual MEFs quarterly MEFs FWPs 8-Ks and earnings announcements within 30 days of

the SEO issue date

8-K abnormal size The SEO firmrsquos 8-K size less the median non-SEO 8-K size

8-K size The size in megabytes of the last 8-K during the 30 days prior to the SEO issue date otherwise set to zero

32

References

Aboody D Kasznik R 2000 CEO stock option awards and the timing of voluntary corporate

disclosures J of Account and Econ 29 (1) 73ndash100

Ajinkya B Bhojraj S Sengupta P 2005 The association between outside directors institutional

investors and the properties of management earnings forecasts J of Account Res 43 (3) 343ndash 376

Aldrich JH Nelson FD 1984 Linear Probability Logit and Probit Models SAGE Publications

Beverly Hills CA

Ali A Klasa S Yeung E 2014 Industry concentration and corporate disclosure policy Working

paper The University of Texas at Dallas

Asquith P Mullins DW 1986 Equity issues and offering dilution J of Financ Econ 15 (1) 61ndash89

Autore DM Kumar R Shome DK 2008 The revival of shelf-registered corporate equity offerings

J of Corp Financ 14 (1) 32ndash50

Bailey W Karolyi GA Salva C 2006 The economic consequences of increased disclosure Evidence

from international cross-listings J of Financ Econ 81 (1) 175ndash213

Beck T Levine R Levkov A 2010 Big bad banks The winners and losers from bank deregulation in

the United States J of Financ 65 (5) 1637ndash1667

Bethel JE Krigman L 2008 Managing the costs of issuing common equity The role of registration

choice Q J of Financ and Account 47 (4) 57ndash85

Beyer A Cohen DA Lys TZ Walther BR 2010 The financial reporting environment Review of

the recent literature J of Account and Econ 50 (2ndash3) 179ndash234

Blackwell DW Marr MW Spivey MF 1990 Shelf registration and the reduced due diligence

argument Implications of the underwriter certification and the implicit insurance hypotheses J

of Financ and Quant Anal 25 (2) 245ndash259

Botosan CA 1997 Disclosure level and the cost of equity capital Account Rev 72 (3) 323ndash349

Brockman P Khurana IK Martin X 2008 Voluntary disclosures around share repurchases J of

Financ Econ 89 (1) 175ndash191

Cadwalader Wickersham amp Taft LLP 2005 New SEC rules for Securities Offering Reform August 4

httpwwwcadwaladercomassetsclient_friend080405SECRulesforSecuritiesOffering

Reformpdf [Accessed 16 Dec 2011]

Chang X Dasgupta S Hillary G 2006 Analyst coverage and financing decisions J of Financ 61 (6)

2009ndash2048

Cheng Q Lo K 2006 Insider trading and voluntary disclosures J of Account Res 44 (5) 815ndash848

33

Appendix Variable definitions and other key terms

Variable Definition

Abnormal magnitude

Abnormal return

Analyst dispersion

Analyst following

Annual MEF

Bias

Book-to-market

Closest disclosure

Distance

Earnings announcement

Error

Firm risk

First annual MEF

First quarterly MEF

Frequency

Forecasting

FWP

Guidance

Horizon

Institutional ownership

Last annual MEF

Last MEF

The magnitude over trading days [-200] less the magnitude over trading days [+1+20] The return over trading days [-200] less the return over trading days [+1+20] The standard deviation of analystsrsquo annual earnings estimates divided by the absolute value of the median analyst earnings forecast Analyst dispersion is calculated using the last reported analyst estimates in Institutional Brokersrsquo Estimate System (IBES) for the month-end at least 30 days prior to the seasoned equity offering (SEO) issue date The number of analysts providing annual forecasts at the month ending at least 30 days prior to the SEO issue date A firm-issued point or median of the range forecast of annual earnings The firmrsquos annual MEF or quarterly MEF minus the actual earnings divided by the stock price at the beginning of the fiscal period annualized The book value of common equity divided by the market value of common equity at the fiscal year-end prior to the SEO issue date Book-to-market is set to the minimum sample value for two observations with missing values The number of days between the last disclosure of the following type prior to the SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement The number of days between the specified disclosure and the SEO issue date The periodic announcement of quarterly or annual earnings during the 30 days prior to the SEO issue date The absolute value of the firmrsquos annual MEF or quarterly MEF minus the actual earnings divided by the stock price at the beginning of the fiscal period annualized The standard deviation of the firms daily returns over the period [-250-20] trading days to the SEO issue date The annual MEF issued farthest from the SEO within 365 days of the SEO issue date The quarterly MEF issued farthest from the SEO within 365 days of the SEO issue date The number of the specified disclosures during the month prior to the SEO issue date An indicator variable equal to one if the firm provides an annual MEF or quarterly MEF during the 365 days prior to the SEO issue date A free writing prospectus is an SEC filing of communication FWPs can include any written communication in printed form and other communication via television radio broadcasts and certain electronic road shows (Pena 2007) A forward-looking disclosure in the form of an MEF or FWP The number of days between the annual MEF or quarterly MEF and the fiscal period end annualized The percentage of the firmrsquos outstanding shares held by institutions at the quarter-end prior to the SEO issue date The annual MEF issued closest to the SEO within 30 days of the SEO issue date The last quarterly MEF or annual MEF within 30 days of the SEO issue date

31

Last quarterly MEF The quarterly MEF issued closest to the SEO within 30 days of the SEO issue date

Litigation risk An indicator variable equal to one if the SEO firm operates in the following high litigation risk industries (standard

industrial classification codes) biotechnology (2833ndash2836) research and development services (8731ndash8734)

programming (7371ndash7379) computers (3570ndash3577) electronics (3600ndash3674) or retailing (5200ndash5961)

Ln(market capitalization) The natural log of the market value of common equity at fiscal year-end prior to the SEO issue date

Loss An indicator variable equal to one if the firm reports a loss in the current fiscal year

Magnitude The absolute value of the daily return net of the return on the CRSP value-weighted index cumulated over the

specified trading days [-200] and [+0+20] with day 0 being the volume-based corrected issue date (Corwin 2003)

Market capitalization The market value of common equity at the fiscal year-end prior to the SEO issue date in billions of dollars

MEF A management earnings forecast a firm-issued point or median of range forecast of annual or quarterly earnings

Non-forecasting An indicator variable equal to one if the firm does not provide an annual MEF or quarterly MEF during the 365 days

Non-SEO 8-K size

prior to the SEO issue date

The median 8-K size for non-SEO well-known seasoned issuers (WKSIs) during the same period Non-SEO WKSIs

meet the WKSI market capitalization criteria but do not conduct an SEO during the same calendar year

Occurrence An indicator variable equal to one if the firm issues the specified voluntary disclosure (annual MEF quarterly MEF

or FWP) within 30 days prior to the SEO issue date

Pre-issue days The number of days between the filing of a preliminary prospectus (pre-issue date) and the SEO issue date

Quarterly MEF A firm-issued point or median of range forecast of quarterly earnings

Registration days The number of days between the filing of a registration statement and the SEO issue date

Residual analyst following The residual value after regressing analyst following on Ln(market capitalization)

Return

SEO proceeds

SEO size

Shelf

Short distance 8-K

Short distance disclosure

SOR

Total disclosure

The raw daily return cumulated over the specified trading days [-200] and [+0+20] with day 0 being the volume-

based corrected issue date (Corwin 2003)

The gross proceeds from the seasoned equity offering before underwriting

The SEO proceeds scaled by market capitalization at the fiscal year-end prior to the SEO issue date

An indicator variable equal to one if the SEO is offered from a shelf registration

An indicator variable equal to one if the firm provides an 8-K during the five days prior to the SEO issue date

An indicator variable equal to one if the firm provides any of the following disclosures during the five days prior to the

SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement

An indicator variable equal to one if the SEO issue date is on or after December 1 2005 when SOR became effective

The aggregate number of annual MEFs quarterly MEFs FWPs 8-Ks and earnings announcements within 30 days of

the SEO issue date

8-K abnormal size The SEO firmrsquos 8-K size less the median non-SEO 8-K size

8-K size The size in megabytes of the last 8-K during the 30 days prior to the SEO issue date otherwise set to zero

32

References

Aboody D Kasznik R 2000 CEO stock option awards and the timing of voluntary corporate

disclosures J of Account and Econ 29 (1) 73ndash100

Ajinkya B Bhojraj S Sengupta P 2005 The association between outside directors institutional

investors and the properties of management earnings forecasts J of Account Res 43 (3) 343ndash 376

Aldrich JH Nelson FD 1984 Linear Probability Logit and Probit Models SAGE Publications

Beverly Hills CA

Ali A Klasa S Yeung E 2014 Industry concentration and corporate disclosure policy Working

paper The University of Texas at Dallas

Asquith P Mullins DW 1986 Equity issues and offering dilution J of Financ Econ 15 (1) 61ndash89

Autore DM Kumar R Shome DK 2008 The revival of shelf-registered corporate equity offerings

J of Corp Financ 14 (1) 32ndash50

Bailey W Karolyi GA Salva C 2006 The economic consequences of increased disclosure Evidence

from international cross-listings J of Financ Econ 81 (1) 175ndash213

Beck T Levine R Levkov A 2010 Big bad banks The winners and losers from bank deregulation in

the United States J of Financ 65 (5) 1637ndash1667

Bethel JE Krigman L 2008 Managing the costs of issuing common equity The role of registration

choice Q J of Financ and Account 47 (4) 57ndash85

Beyer A Cohen DA Lys TZ Walther BR 2010 The financial reporting environment Review of

the recent literature J of Account and Econ 50 (2ndash3) 179ndash234

Blackwell DW Marr MW Spivey MF 1990 Shelf registration and the reduced due diligence

argument Implications of the underwriter certification and the implicit insurance hypotheses J

of Financ and Quant Anal 25 (2) 245ndash259

Botosan CA 1997 Disclosure level and the cost of equity capital Account Rev 72 (3) 323ndash349

Brockman P Khurana IK Martin X 2008 Voluntary disclosures around share repurchases J of

Financ Econ 89 (1) 175ndash191

Cadwalader Wickersham amp Taft LLP 2005 New SEC rules for Securities Offering Reform August 4

httpwwwcadwaladercomassetsclient_friend080405SECRulesforSecuritiesOffering

Reformpdf [Accessed 16 Dec 2011]

Chang X Dasgupta S Hillary G 2006 Analyst coverage and financing decisions J of Financ 61 (6)

2009ndash2048

Cheng Q Lo K 2006 Insider trading and voluntary disclosures J of Account Res 44 (5) 815ndash848

33

Last quarterly MEF The quarterly MEF issued closest to the SEO within 30 days of the SEO issue date

Litigation risk An indicator variable equal to one if the SEO firm operates in the following high litigation risk industries (standard

industrial classification codes) biotechnology (2833ndash2836) research and development services (8731ndash8734)

programming (7371ndash7379) computers (3570ndash3577) electronics (3600ndash3674) or retailing (5200ndash5961)

Ln(market capitalization) The natural log of the market value of common equity at fiscal year-end prior to the SEO issue date

Loss An indicator variable equal to one if the firm reports a loss in the current fiscal year

Magnitude The absolute value of the daily return net of the return on the CRSP value-weighted index cumulated over the

specified trading days [-200] and [+0+20] with day 0 being the volume-based corrected issue date (Corwin 2003)

Market capitalization The market value of common equity at the fiscal year-end prior to the SEO issue date in billions of dollars

MEF A management earnings forecast a firm-issued point or median of range forecast of annual or quarterly earnings

Non-forecasting An indicator variable equal to one if the firm does not provide an annual MEF or quarterly MEF during the 365 days

Non-SEO 8-K size

prior to the SEO issue date

The median 8-K size for non-SEO well-known seasoned issuers (WKSIs) during the same period Non-SEO WKSIs

meet the WKSI market capitalization criteria but do not conduct an SEO during the same calendar year

Occurrence An indicator variable equal to one if the firm issues the specified voluntary disclosure (annual MEF quarterly MEF

or FWP) within 30 days prior to the SEO issue date

Pre-issue days The number of days between the filing of a preliminary prospectus (pre-issue date) and the SEO issue date

Quarterly MEF A firm-issued point or median of range forecast of quarterly earnings

Registration days The number of days between the filing of a registration statement and the SEO issue date

Residual analyst following The residual value after regressing analyst following on Ln(market capitalization)

Return

SEO proceeds

SEO size

Shelf

Short distance 8-K

Short distance disclosure

SOR

Total disclosure

The raw daily return cumulated over the specified trading days [-200] and [+0+20] with day 0 being the volume-

based corrected issue date (Corwin 2003)

The gross proceeds from the seasoned equity offering before underwriting

The SEO proceeds scaled by market capitalization at the fiscal year-end prior to the SEO issue date

An indicator variable equal to one if the SEO is offered from a shelf registration

An indicator variable equal to one if the firm provides an 8-K during the five days prior to the SEO issue date

An indicator variable equal to one if the firm provides any of the following disclosures during the five days prior to the

SEO issue date annual MEF quarterly MEF FWP 8-K or earnings announcement

An indicator variable equal to one if the SEO issue date is on or after December 1 2005 when SOR became effective

The aggregate number of annual MEFs quarterly MEFs FWPs 8-Ks and earnings announcements within 30 days of

the SEO issue date

8-K abnormal size The SEO firmrsquos 8-K size less the median non-SEO 8-K size

8-K size The size in megabytes of the last 8-K during the 30 days prior to the SEO issue date otherwise set to zero

32

References

Aboody D Kasznik R 2000 CEO stock option awards and the timing of voluntary corporate

disclosures J of Account and Econ 29 (1) 73ndash100

Ajinkya B Bhojraj S Sengupta P 2005 The association between outside directors institutional

investors and the properties of management earnings forecasts J of Account Res 43 (3) 343ndash 376

Aldrich JH Nelson FD 1984 Linear Probability Logit and Probit Models SAGE Publications

Beverly Hills CA

Ali A Klasa S Yeung E 2014 Industry concentration and corporate disclosure policy Working

paper The University of Texas at Dallas

Asquith P Mullins DW 1986 Equity issues and offering dilution J of Financ Econ 15 (1) 61ndash89

Autore DM Kumar R Shome DK 2008 The revival of shelf-registered corporate equity offerings

J of Corp Financ 14 (1) 32ndash50

Bailey W Karolyi GA Salva C 2006 The economic consequences of increased disclosure Evidence

from international cross-listings J of Financ Econ 81 (1) 175ndash213

Beck T Levine R Levkov A 2010 Big bad banks The winners and losers from bank deregulation in

the United States J of Financ 65 (5) 1637ndash1667

Bethel JE Krigman L 2008 Managing the costs of issuing common equity The role of registration

choice Q J of Financ and Account 47 (4) 57ndash85

Beyer A Cohen DA Lys TZ Walther BR 2010 The financial reporting environment Review of

the recent literature J of Account and Econ 50 (2ndash3) 179ndash234

Blackwell DW Marr MW Spivey MF 1990 Shelf registration and the reduced due diligence

argument Implications of the underwriter certification and the implicit insurance hypotheses J

of Financ and Quant Anal 25 (2) 245ndash259

Botosan CA 1997 Disclosure level and the cost of equity capital Account Rev 72 (3) 323ndash349

Brockman P Khurana IK Martin X 2008 Voluntary disclosures around share repurchases J of

Financ Econ 89 (1) 175ndash191

Cadwalader Wickersham amp Taft LLP 2005 New SEC rules for Securities Offering Reform August 4

httpwwwcadwaladercomassetsclient_friend080405SECRulesforSecuritiesOffering

Reformpdf [Accessed 16 Dec 2011]

Chang X Dasgupta S Hillary G 2006 Analyst coverage and financing decisions J of Financ 61 (6)

2009ndash2048

Cheng Q Lo K 2006 Insider trading and voluntary disclosures J of Account Res 44 (5) 815ndash848

33

References

Aboody D Kasznik R 2000 CEO stock option awards and the timing of voluntary corporate

disclosures J of Account and Econ 29 (1) 73ndash100

Ajinkya B Bhojraj S Sengupta P 2005 The association between outside directors institutional

investors and the properties of management earnings forecasts J of Account Res 43 (3) 343ndash 376

Aldrich JH Nelson FD 1984 Linear Probability Logit and Probit Models SAGE Publications

Beverly Hills CA

Ali A Klasa S Yeung E 2014 Industry concentration and corporate disclosure policy Working

paper The University of Texas at Dallas

Asquith P Mullins DW 1986 Equity issues and offering dilution J of Financ Econ 15 (1) 61ndash89

Autore DM Kumar R Shome DK 2008 The revival of shelf-registered corporate equity offerings

J of Corp Financ 14 (1) 32ndash50

Bailey W Karolyi GA Salva C 2006 The economic consequences of increased disclosure Evidence

from international cross-listings J of Financ Econ 81 (1) 175ndash213

Beck T Levine R Levkov A 2010 Big bad banks The winners and losers from bank deregulation in

the United States J of Financ 65 (5) 1637ndash1667

Bethel JE Krigman L 2008 Managing the costs of issuing common equity The role of registration

choice Q J of Financ and Account 47 (4) 57ndash85

Beyer A Cohen DA Lys TZ Walther BR 2010 The financial reporting environment Review of

the recent literature J of Account and Econ 50 (2ndash3) 179ndash234

Blackwell DW Marr MW Spivey MF 1990 Shelf registration and the reduced due diligence

argument Implications of the underwriter certification and the implicit insurance hypotheses J

of Financ and Quant Anal 25 (2) 245ndash259

Botosan CA 1997 Disclosure level and the cost of equity capital Account Rev 72 (3) 323ndash349

Brockman P Khurana IK Martin X 2008 Voluntary disclosures around share repurchases J of

Financ Econ 89 (1) 175ndash191

Cadwalader Wickersham amp Taft LLP 2005 New SEC rules for Securities Offering Reform August 4

httpwwwcadwaladercomassetsclient_friend080405SECRulesforSecuritiesOffering

Reformpdf [Accessed 16 Dec 2011]

Chang X Dasgupta S Hillary G 2006 Analyst coverage and financing decisions J of Financ 61 (6)

2009ndash2048

Cheng Q Lo K 2006 Insider trading and voluntary disclosures J of Account Res 44 (5) 815ndash848

33

Chuk E Matsumoto D Miller G 2013 Assessing methods of identifying management forecasts CIG

vs researcher collected J of Account and Econ 55 (1) 23ndash42

Coffee JC 2005 A Section 11 safe harbor NY Law J 234 5ndash9

Cohen DA Zarowin P 2010 Accrual-based and real earnings management activities around seasoned

equity offerings J of Account and Econ 50 (1) 2ndash19

Corwin SA 2003 The determinants of underpricing for seasoned equity offers J of Financ 58 (5)

2249ndash2279

Diamond DW Verrecchia RE 1991 Disclosure liquidity and the cost of capital J of Financ 46

(4) 1325ndash1359

Drake MS Roulstone DT Thornock JR 2014 The determinants and consequences of information

acquisition via EDGAR Contemp Account Res forthcoming

Fernandes N Lei U Miller DP 2010 Escape from New York The market impact of loosening

disclosure requirements J of Financ Econ 95 (2) 129ndash147

Field L Lowry M Shu S 2005 Does disclosure deter or trigger litigation J of Account and Econ

39 (3) 487ndash507

Francis J Schipper K Vincent L 2002 Earnings announcements and competing information J of

Account and Econ 33 (3) 313ndash342

Gong G Li LY Xie H 2009 The association between management earnings forecasts and accruals

Account Rev 84 (2) 497ndash530

Graham JR Harvey CR Rajgopal S 2005 The economic implications of corporate financial

reporting J of Account and Econ 40 (1ndash3) 3ndash73

Gu Z Chen T 2004 Analystsrsquo treatment of nonrecurring items in street earnings J of Account and Econ 38 129ndash170

Hanley KW Hoberg G 2013 The information content of IPO prospectuses Rev of Financ Stud 23

(7) 2821ndash2864

Henry TR Koski JL 2010 Short selling around seasoned equity offerings Rev of Financ Stud 23

(12) 4389ndash4418

Kim I Skinner DJ 2012 Measuring securities litigation risk J of Account and Econ 53 (1ndash2) 290ndash 310

Korajczyk RA Lucas DJ McDonald RL 1991 The effect of information releases on the pricing

and timing of equity issues Rev of Financ Stud 4 (4) 685ndash708

Kothari SP Shu S Wysocki PD 2009 Do managers withhold bad news J of Account Res 47 (1)

241ndash276

34

Lang MH Lundholm RJ 2000 Voluntary disclosure and equity offerings Reducing information

asymmetry or hyping the stock Contemp Account Res 17 (4) 623ndash662

Latham amp Watkins LLP 2005 Christmas in JulyndashThe SEC improves the securities offering process

August httpwwwlwcomuploadpubcontent_pdfpub1360_1pdf [Accessed 25 Jun 2013]

Lerman A Livnat J 2010 The new Form 8-K disclosures Rev of Account Stud 15 (4) 752ndash778

Li OZ Zhuang Z 2012 Management guidance and the underpricing of seasoned equity offerings

Contemp Account Res 29 (3) 710ndash737

Marquardt CA Wiedman CI 1998 Voluntary disclosure information asymmetry and insider selling

through secondary equity offerings Contemp Account Res 15 (4) 505ndash537

Masulis RW Korwar AN 1986 Seasoned equity offerings An empirical investigation J of Financ

Econ 15 (1) 91ndash118

Morrissey JF 2007 Rhetoric and reality Investor protection and the securities regulation reform of

2005 Cathol Univ Law Rev 56 (2) 561ndash607

Pena A 2007 The free writing prospectus A six-question approach for issuers Secur Regul Law J 35

(1) 36ndash70

Rogers JL Van Buskirk A 2009 Shareholder litigation and changes in disclosure behavior J of

Account and Econ 47 (1ndash2) 136ndash156

Ruland W Tung S George NE 1990 Factors associated with the disclosure of managersrsquo forecasts

Account Rev 65 (3) 710ndash721

Segal B Segal D 2013 The opportunistic reporting of material events and the apparent misconception

of investorsrsquo reaction Working paper INSEAD

Shroff NO Sun AX White HD Zhang W 2013 Voluntary disclosure and information

asymmetry Evidence from the 2005 Securities Offering Reform J of Account Res 51 (5)

1261ndash1345

Skinner DJ 1994 Why firms voluntarily disclose bad news J of Account Res 32 (1) 38ndash60

Tetlock PC 2010 Does public financial news resolve asymmetric information Rev of Financ Stud

23 (9) 3520ndash3557

35

Lang MH Lundholm RJ 2000 Voluntary disclosure and equity offerings Reducing information

asymmetry or hyping the stock Contemp Account Res 17 (4) 623ndash662

Latham amp Watkins LLP 2005 Christmas in JulyndashThe SEC improves the securities offering process

August httpwwwlwcomuploadpubcontent_pdfpub1360_1pdf [Accessed 25 Jun 2013]

Lerman A Livnat J 2010 The new Form 8-K disclosures Rev of Account Stud 15 (4) 752ndash778

Li OZ Zhuang Z 2012 Management guidance and the underpricing of seasoned equity offerings

Contemp Account Res 29 (3) 710ndash737

Marquardt CA Wiedman CI 1998 Voluntary disclosure information asymmetry and insider selling

through secondary equity offerings Contemp Account Res 15 (4) 505ndash537

Masulis RW Korwar AN 1986 Seasoned equity offerings An empirical investigation J of Financ

Econ 15 (1) 91ndash118

Morrissey JF 2007 Rhetoric and reality Investor protection and the securities regulation reform of

2005 Cathol Univ Law Rev 56 (2) 561ndash607

Pena A 2007 The free writing prospectus A six-question approach for issuers Secur Regul Law J 35

(1) 36ndash70

Rogers JL Van Buskirk A 2009 Shareholder litigation and changes in disclosure behavior J of

Account and Econ 47 (1ndash2) 136ndash156

Ruland W Tung S George NE 1990 Factors associated with the disclosure of managersrsquo forecasts

Account Rev 65 (3) 710ndash721

Segal B Segal D 2013 The opportunistic reporting of material events and the apparent misconception

of investorsrsquo reaction Working paper INSEAD

Shroff NO Sun AX White HD Zhang W 2013 Voluntary disclosure and information

asymmetry Evidence from the 2005 Securities Offering Reform J of Account Res 51 (5)

1261ndash1345

Skinner DJ 1994 Why firms voluntarily disclose bad news J of Account Res 32 (1) 38ndash60

Tetlock PC 2010 Does public financial news resolve asymmetric information Rev of Financ Stud

23 (9) 3520ndash3557

35

Fig 1 Timeline of shelf registration and issue before and after the Securities Offering

Reform (SOR)

Pre-SOR

Post-SOR

Denotes changes from pre-SOR

36

Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)

100

90

Cu

mu

lati

ve

ma

gn

itu

de

(per

cen

t)

80

70

60

50

40

30

20

10

0

-20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0

Trading days prior to SEO issue date

Pre-SOR Post-SOR

This figure presents the cumulative magnitude beginning 20 trading days prior to the SEO issue date

before and after the Securities Offering Reform (SOR) Magnitude is calculated as the absolute value of

daily return net of the return on the CRSP value-weighted index

37

Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity

offering (SEO)

20

18

Ag

gre

ga

te m

ag

nit

ud

e (p

ercen

t)

16

14

12

10

8

6

4

2

0

Trading days prior to SEO issue date

Pre-SOR Post-SOR

-20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0

The figure plots three-day [-1+1] magnitude aggregated for all 8-K filing dates beginning 20 trading days

prior to the SEO issue date before and after the Securities Offering Reform (SOR) Three-day magnitude

is calculated as the cumulative three-day [-1+1] absolute value of daily returns net of the returns on the

CRSP value-weighted index

38

Table 1 Sample distribution and offering characteristics

Panel A Distribution of SEOs by calendar year

Full

Pre- Post- Sample

Sample 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Forecasting 16 15 15 12 17 15 12 24 55 71 126

Non-forecasting 14 23 30 26 19 31 36 55 90 144 234

Full Sample 30 38 45 38 36 46 48 79 145 215 360

Panel B Offering characteristics for SEOs by calendar year

Full

Pre- Post- Sample

Variable 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Shelf (percent) 500 763 822 895 861 826 917 861 752 870 822

Registration days 621 1878 2136 2661 2455 1647 3926 3618 1960 2984 2571

Pre-issue days 34 37 35 26 18 38 20 11 33 20 26

SEO proceeds (millions of dollars) 2372 2232 3564 3400 2805 3508 6239 3235 2872 3937 3508

SEO size (percent) 106 137 127 123 114 140 98 131 123 124 123

Panel C Average number of 8-Ks filed during SEO fiscal year

Full

Pre- Post- Sample

Sample 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Forecasting 64 128 133 194 159 141 166 170 128 160 144 Non-forecasting 99 135 143 223 193 189 190 190 155 190 177

Full Sample 81 130 144 216 177 170 185 183 146 179 165

This table presents the distribution values for our full sample of 360 seasoned equity offerings (SEOs) between 2002 and 2009 The Securities

Offering Reform (SOR) became effective December 1 2005 Panels A and B are presented at the SEO level and Panel C is presented at the issuer-

year level We designate SEOs issued between January 1 2002 and November 30 2005 as pre-SOR and SEOs issued between December 1 2005

and December 31 2009 as post-SOR Three SEOs in the 2005 forecasting sample and three SEOs in the 2005 non-forecasting sample occur in

December 2005 These SEOs are included in the 2005 and post-SOR columns The final column presents the total number in Panel A and the

average values in Panels B and C for the full sample period The forecasting (non-forecasting) sample contains SEOs in which the firm has chosen

(not) to provide at least one annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue

date All variables are defined in the Appendix

39

Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Market capitalization (billions of dollars)

Institutional ownership (percent)

522506

6180

601371

7262

888659

6709

850331

7500

298746

5856

478620

7144

Analyst following

Analyst dispersion (percent)

Litigation risk (percent)

Book-to-market

1039

1408

3172

038

1087

2218

2233

044

935

1446

2545

041

1011

797

2254

044

1102

1384

3556

036

1122

2919

2222

044

Loss (percent)

Firm risk (percent)

2897

252

3535

376

1636

273

1690

323

3667

280

4444

402

This table presents the mean values of descriptive statistics for our control variables The forecasting (non-forecasting) sample contains seasoned

equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or quarterly point or range management earnings forecast

(MEF) or both within one year before the SEO issue date The full sample contains both the forecasting and non-forecasting samples and

indicate that the difference in means is significant at the 1 5 and 10 level respectively using a standard two-tailed t-test All variables are

defined in the Appendix

40

Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N= 215) (N = 55) (N = 71) (N = 90) (N = 144)

Panel A Occurrence of guidance disclosure during one month prior to SEO

Annual MEF 1034 1488 2727 4507

mdash mdash Quarterly MEF 1241 1163 3273 3531 mdash mdash All MEF 1724 1953 4545 5915 mdash mdash FWP NA 930 NA 1268 NA 764

Guidance 1724 2605 4545 6338 mdash 764

Panel B Frequency of disclosure during one month prior to SEO

Total disclosure 246 308

271 358

231 283

Annual MEF 012 020 031 059

mdash mdash Quarterly MEF 014 017 038 051 mdash mdash All MEF 026 036 069 110

mdash mdash

FWP NA 011 NA 013 NA 010

Guidance 026 047

069 123

mdash 010

Earnings announcement 047 048 044 044 049 051

8-K 173 213

158 192

182 223

Panel C Size of Form 8-K disclosure during one month prior to SEO

8-K size 254 597

297 830

229 481

8-K abnormal size 221 517

267 754 194 399

Panel D Distance between last disclosure and SEO issue date for prior one-month

Annual MEF mdash mdash 1473 1188 mdash mdash Quarterly MEF mdash mdash 1494 904

mdash mdash

All MEF mdash mdash 1456 1052

mdash mdash FWP NA 315 NA 278 NA 345

Guidance mdash mdash 1456 971

mdash mdash Earnings announcement 1440 1346 1329 1439 1500 1307

8-K 813 684 860 837 783 605

Closest disclosure 785 615

824 672 760 586

This table presents mean values for multiple forms of disclosure before and after SOR The forecasting (non-forecasting)

sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The

full sample contains both the forecasting and non-forecasting samples Panel D is for the subgroup of SEO firms with the

respective type of disclosure in the prior one-month period and indicate significance at the 1 5 and 10

level respectively using a standard two-tailed t-test NA = not applicable All variables are defined in the Appendix

41

Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)

Panel A Occurrence frequency and 8-K size during one-month period prior to SEO issue date

Variable

Occurrence of

annual MEF

Occurrence of

guidance

Frequency of

guidance

Frequency of

8-Ks

Frequency of

total disclosure

File size of

last 8-K

File size of

last 8-K

Constant

(1) -189

(2) -125

(3) -002

(4) 085

(5) 154

(6) -1216

(7) -1223

(005) (018) (098) (014) (005) (021) (020)

SOR 043

047

048

038

055

198

492

(009) (006) (004) (001) (001) (006) (004)

Ln(market capitalization) 010 -002 -002 014

010 151 173

(032) (083) (087) (009) (034) (013) (011)

Institutional ownership 110

113

107

-019 036 235 189

(004) (001) (001) (045) (030) (031) (039)

Residual analyst following -002 -006

-004

004

002 001 002

(046) (004) (005) (010) (046) (095) (091)

Analyst dispersion -474

076 038 -011 -014 515 497

(006) (019) (029) (051) (060) (034) (035)

Litigation risk -020 025 -002 -017 -010 -114 -106

(051) (041) (091) (032) (067) (036) (042)

Book-to-market 018 071

024 -015 -001 327 363

(047) (003) (010) (031) (098) (030) (027)

Loss -036 -068

-041

008 -027 -222

-220

(034) (007) (008) (062) (019) (004) (005)

Firm risk 156 391 041 159 340 3324 6409

(086) (064) (094) (065) (049) (024) (011)

Non-SEO 8-K size -690

(014)

Sample Forecasting Forecasting Forecasting Full Full Full Full

Regression Probit Probit OLS OLS OLS OLS OLS

Observations 126 126 126 360 360 358 358

Pseudo R2

0150 0120

R2

010 006 004 006 008

Adjusted R2

003 003 002 004 005

42

Table 4 (continued)

Panel B Distance between last disclosure and SEO issue date by disclosure type

Variable

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Constant

(1) 2456

(2) 1595

(3) 1647

(4) 10149

(5) 1684

(6) 1690

(000) (000) (000) (019) (008) (001)

SOR -563

-083 -106 -1693 -691

-372

(003) (033) (020) (027) (002) (004)

Ln(market capitalization) -112 -104

-102

509 017 -043

(018) (000) (000) (059) (088) (060)

Institutional ownership 350 082 002 -5322

206 -017

(046) (058) (099) (008) (059) (095)

Residual analyst following 001 -023

-018

299 -046

-029

(096) (001) (003) (021) (002) (006)

Analyst dispersion -162 -053 -035 -3270 -157 -156

(064) (060) (070) (018) (053) (041)

Litigation risk -087 020 027 -668 152 216

(070) (083) (076) (073) (062) (026)

Book-to-market 023 034 -031 -633 -372 042

(085) (069) (065) (048) (045) (073)

Loss -020 -096 -085 5399

164 226

(096) (035) (037) (005) (049) (025)

Firm risk -11517

-1358 -2299 -69596 -4750 -6360

(006) (052) (023) (011) (024) (004)

Restricted to one month Yes Yes Yes No No No

Regression OLS OLS OLS OLS OLS OLS

Observations 81 322 332 146 358 360

R2

018 006 006 010 006 005

Adjusted R2

008 003 004 004 003 003

In Panel A the forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The full sample contains both the forecasting

and non-forecasting samples In Panel B Columns 1ndash3 (4ndash6) contain SEOs in which the firm provides a specified disclosure within one month (one year) before

the SEO issue date p-values in parentheses are based on robust standard errors clustered at the firm level and indicate significance at the 1 5 and

10 level respectively from a two-tailed test All variables are defined in the Appendix

43

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)

100

90

Cu

mu

lati

ve

ma

gn

itu

de

(per

cen

t)

80

70

60

50

40

30

20

10

0

-20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0

Trading days prior to SEO issue date

Pre-SOR Post-SOR

This figure presents the cumulative magnitude beginning 20 trading days prior to the SEO issue date

before and after the Securities Offering Reform (SOR) Magnitude is calculated as the absolute value of

daily return net of the return on the CRSP value-weighted index

37

Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity

offering (SEO)

20

18

Ag

gre

ga

te m

ag

nit

ud

e (p

ercen

t)

16

14

12

10

8

6

4

2

0

Trading days prior to SEO issue date

Pre-SOR Post-SOR

-20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0

The figure plots three-day [-1+1] magnitude aggregated for all 8-K filing dates beginning 20 trading days

prior to the SEO issue date before and after the Securities Offering Reform (SOR) Three-day magnitude

is calculated as the cumulative three-day [-1+1] absolute value of daily returns net of the returns on the

CRSP value-weighted index

38

Table 1 Sample distribution and offering characteristics

Panel A Distribution of SEOs by calendar year

Full

Pre- Post- Sample

Sample 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Forecasting 16 15 15 12 17 15 12 24 55 71 126

Non-forecasting 14 23 30 26 19 31 36 55 90 144 234

Full Sample 30 38 45 38 36 46 48 79 145 215 360

Panel B Offering characteristics for SEOs by calendar year

Full

Pre- Post- Sample

Variable 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Shelf (percent) 500 763 822 895 861 826 917 861 752 870 822

Registration days 621 1878 2136 2661 2455 1647 3926 3618 1960 2984 2571

Pre-issue days 34 37 35 26 18 38 20 11 33 20 26

SEO proceeds (millions of dollars) 2372 2232 3564 3400 2805 3508 6239 3235 2872 3937 3508

SEO size (percent) 106 137 127 123 114 140 98 131 123 124 123

Panel C Average number of 8-Ks filed during SEO fiscal year

Full

Pre- Post- Sample

Sample 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Forecasting 64 128 133 194 159 141 166 170 128 160 144 Non-forecasting 99 135 143 223 193 189 190 190 155 190 177

Full Sample 81 130 144 216 177 170 185 183 146 179 165

This table presents the distribution values for our full sample of 360 seasoned equity offerings (SEOs) between 2002 and 2009 The Securities

Offering Reform (SOR) became effective December 1 2005 Panels A and B are presented at the SEO level and Panel C is presented at the issuer-

year level We designate SEOs issued between January 1 2002 and November 30 2005 as pre-SOR and SEOs issued between December 1 2005

and December 31 2009 as post-SOR Three SEOs in the 2005 forecasting sample and three SEOs in the 2005 non-forecasting sample occur in

December 2005 These SEOs are included in the 2005 and post-SOR columns The final column presents the total number in Panel A and the

average values in Panels B and C for the full sample period The forecasting (non-forecasting) sample contains SEOs in which the firm has chosen

(not) to provide at least one annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue

date All variables are defined in the Appendix

39

Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Market capitalization (billions of dollars)

Institutional ownership (percent)

522506

6180

601371

7262

888659

6709

850331

7500

298746

5856

478620

7144

Analyst following

Analyst dispersion (percent)

Litigation risk (percent)

Book-to-market

1039

1408

3172

038

1087

2218

2233

044

935

1446

2545

041

1011

797

2254

044

1102

1384

3556

036

1122

2919

2222

044

Loss (percent)

Firm risk (percent)

2897

252

3535

376

1636

273

1690

323

3667

280

4444

402

This table presents the mean values of descriptive statistics for our control variables The forecasting (non-forecasting) sample contains seasoned

equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or quarterly point or range management earnings forecast

(MEF) or both within one year before the SEO issue date The full sample contains both the forecasting and non-forecasting samples and

indicate that the difference in means is significant at the 1 5 and 10 level respectively using a standard two-tailed t-test All variables are

defined in the Appendix

40

Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N= 215) (N = 55) (N = 71) (N = 90) (N = 144)

Panel A Occurrence of guidance disclosure during one month prior to SEO

Annual MEF 1034 1488 2727 4507

mdash mdash Quarterly MEF 1241 1163 3273 3531 mdash mdash All MEF 1724 1953 4545 5915 mdash mdash FWP NA 930 NA 1268 NA 764

Guidance 1724 2605 4545 6338 mdash 764

Panel B Frequency of disclosure during one month prior to SEO

Total disclosure 246 308

271 358

231 283

Annual MEF 012 020 031 059

mdash mdash Quarterly MEF 014 017 038 051 mdash mdash All MEF 026 036 069 110

mdash mdash

FWP NA 011 NA 013 NA 010

Guidance 026 047

069 123

mdash 010

Earnings announcement 047 048 044 044 049 051

8-K 173 213

158 192

182 223

Panel C Size of Form 8-K disclosure during one month prior to SEO

8-K size 254 597

297 830

229 481

8-K abnormal size 221 517

267 754 194 399

Panel D Distance between last disclosure and SEO issue date for prior one-month

Annual MEF mdash mdash 1473 1188 mdash mdash Quarterly MEF mdash mdash 1494 904

mdash mdash

All MEF mdash mdash 1456 1052

mdash mdash FWP NA 315 NA 278 NA 345

Guidance mdash mdash 1456 971

mdash mdash Earnings announcement 1440 1346 1329 1439 1500 1307

8-K 813 684 860 837 783 605

Closest disclosure 785 615

824 672 760 586

This table presents mean values for multiple forms of disclosure before and after SOR The forecasting (non-forecasting)

sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The

full sample contains both the forecasting and non-forecasting samples Panel D is for the subgroup of SEO firms with the

respective type of disclosure in the prior one-month period and indicate significance at the 1 5 and 10

level respectively using a standard two-tailed t-test NA = not applicable All variables are defined in the Appendix

41

Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)

Panel A Occurrence frequency and 8-K size during one-month period prior to SEO issue date

Variable

Occurrence of

annual MEF

Occurrence of

guidance

Frequency of

guidance

Frequency of

8-Ks

Frequency of

total disclosure

File size of

last 8-K

File size of

last 8-K

Constant

(1) -189

(2) -125

(3) -002

(4) 085

(5) 154

(6) -1216

(7) -1223

(005) (018) (098) (014) (005) (021) (020)

SOR 043

047

048

038

055

198

492

(009) (006) (004) (001) (001) (006) (004)

Ln(market capitalization) 010 -002 -002 014

010 151 173

(032) (083) (087) (009) (034) (013) (011)

Institutional ownership 110

113

107

-019 036 235 189

(004) (001) (001) (045) (030) (031) (039)

Residual analyst following -002 -006

-004

004

002 001 002

(046) (004) (005) (010) (046) (095) (091)

Analyst dispersion -474

076 038 -011 -014 515 497

(006) (019) (029) (051) (060) (034) (035)

Litigation risk -020 025 -002 -017 -010 -114 -106

(051) (041) (091) (032) (067) (036) (042)

Book-to-market 018 071

024 -015 -001 327 363

(047) (003) (010) (031) (098) (030) (027)

Loss -036 -068

-041

008 -027 -222

-220

(034) (007) (008) (062) (019) (004) (005)

Firm risk 156 391 041 159 340 3324 6409

(086) (064) (094) (065) (049) (024) (011)

Non-SEO 8-K size -690

(014)

Sample Forecasting Forecasting Forecasting Full Full Full Full

Regression Probit Probit OLS OLS OLS OLS OLS

Observations 126 126 126 360 360 358 358

Pseudo R2

0150 0120

R2

010 006 004 006 008

Adjusted R2

003 003 002 004 005

42

Table 4 (continued)

Panel B Distance between last disclosure and SEO issue date by disclosure type

Variable

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Constant

(1) 2456

(2) 1595

(3) 1647

(4) 10149

(5) 1684

(6) 1690

(000) (000) (000) (019) (008) (001)

SOR -563

-083 -106 -1693 -691

-372

(003) (033) (020) (027) (002) (004)

Ln(market capitalization) -112 -104

-102

509 017 -043

(018) (000) (000) (059) (088) (060)

Institutional ownership 350 082 002 -5322

206 -017

(046) (058) (099) (008) (059) (095)

Residual analyst following 001 -023

-018

299 -046

-029

(096) (001) (003) (021) (002) (006)

Analyst dispersion -162 -053 -035 -3270 -157 -156

(064) (060) (070) (018) (053) (041)

Litigation risk -087 020 027 -668 152 216

(070) (083) (076) (073) (062) (026)

Book-to-market 023 034 -031 -633 -372 042

(085) (069) (065) (048) (045) (073)

Loss -020 -096 -085 5399

164 226

(096) (035) (037) (005) (049) (025)

Firm risk -11517

-1358 -2299 -69596 -4750 -6360

(006) (052) (023) (011) (024) (004)

Restricted to one month Yes Yes Yes No No No

Regression OLS OLS OLS OLS OLS OLS

Observations 81 322 332 146 358 360

R2

018 006 006 010 006 005

Adjusted R2

008 003 004 004 003 003

In Panel A the forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The full sample contains both the forecasting

and non-forecasting samples In Panel B Columns 1ndash3 (4ndash6) contain SEOs in which the firm provides a specified disclosure within one month (one year) before

the SEO issue date p-values in parentheses are based on robust standard errors clustered at the firm level and indicate significance at the 1 5 and

10 level respectively from a two-tailed test All variables are defined in the Appendix

43

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity

offering (SEO)

20

18

Ag

gre

ga

te m

ag

nit

ud

e (p

ercen

t)

16

14

12

10

8

6

4

2

0

Trading days prior to SEO issue date

Pre-SOR Post-SOR

-20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0

The figure plots three-day [-1+1] magnitude aggregated for all 8-K filing dates beginning 20 trading days

prior to the SEO issue date before and after the Securities Offering Reform (SOR) Three-day magnitude

is calculated as the cumulative three-day [-1+1] absolute value of daily returns net of the returns on the

CRSP value-weighted index

38

Table 1 Sample distribution and offering characteristics

Panel A Distribution of SEOs by calendar year

Full

Pre- Post- Sample

Sample 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Forecasting 16 15 15 12 17 15 12 24 55 71 126

Non-forecasting 14 23 30 26 19 31 36 55 90 144 234

Full Sample 30 38 45 38 36 46 48 79 145 215 360

Panel B Offering characteristics for SEOs by calendar year

Full

Pre- Post- Sample

Variable 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Shelf (percent) 500 763 822 895 861 826 917 861 752 870 822

Registration days 621 1878 2136 2661 2455 1647 3926 3618 1960 2984 2571

Pre-issue days 34 37 35 26 18 38 20 11 33 20 26

SEO proceeds (millions of dollars) 2372 2232 3564 3400 2805 3508 6239 3235 2872 3937 3508

SEO size (percent) 106 137 127 123 114 140 98 131 123 124 123

Panel C Average number of 8-Ks filed during SEO fiscal year

Full

Pre- Post- Sample

Sample 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Forecasting 64 128 133 194 159 141 166 170 128 160 144 Non-forecasting 99 135 143 223 193 189 190 190 155 190 177

Full Sample 81 130 144 216 177 170 185 183 146 179 165

This table presents the distribution values for our full sample of 360 seasoned equity offerings (SEOs) between 2002 and 2009 The Securities

Offering Reform (SOR) became effective December 1 2005 Panels A and B are presented at the SEO level and Panel C is presented at the issuer-

year level We designate SEOs issued between January 1 2002 and November 30 2005 as pre-SOR and SEOs issued between December 1 2005

and December 31 2009 as post-SOR Three SEOs in the 2005 forecasting sample and three SEOs in the 2005 non-forecasting sample occur in

December 2005 These SEOs are included in the 2005 and post-SOR columns The final column presents the total number in Panel A and the

average values in Panels B and C for the full sample period The forecasting (non-forecasting) sample contains SEOs in which the firm has chosen

(not) to provide at least one annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue

date All variables are defined in the Appendix

39

Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Market capitalization (billions of dollars)

Institutional ownership (percent)

522506

6180

601371

7262

888659

6709

850331

7500

298746

5856

478620

7144

Analyst following

Analyst dispersion (percent)

Litigation risk (percent)

Book-to-market

1039

1408

3172

038

1087

2218

2233

044

935

1446

2545

041

1011

797

2254

044

1102

1384

3556

036

1122

2919

2222

044

Loss (percent)

Firm risk (percent)

2897

252

3535

376

1636

273

1690

323

3667

280

4444

402

This table presents the mean values of descriptive statistics for our control variables The forecasting (non-forecasting) sample contains seasoned

equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or quarterly point or range management earnings forecast

(MEF) or both within one year before the SEO issue date The full sample contains both the forecasting and non-forecasting samples and

indicate that the difference in means is significant at the 1 5 and 10 level respectively using a standard two-tailed t-test All variables are

defined in the Appendix

40

Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N= 215) (N = 55) (N = 71) (N = 90) (N = 144)

Panel A Occurrence of guidance disclosure during one month prior to SEO

Annual MEF 1034 1488 2727 4507

mdash mdash Quarterly MEF 1241 1163 3273 3531 mdash mdash All MEF 1724 1953 4545 5915 mdash mdash FWP NA 930 NA 1268 NA 764

Guidance 1724 2605 4545 6338 mdash 764

Panel B Frequency of disclosure during one month prior to SEO

Total disclosure 246 308

271 358

231 283

Annual MEF 012 020 031 059

mdash mdash Quarterly MEF 014 017 038 051 mdash mdash All MEF 026 036 069 110

mdash mdash

FWP NA 011 NA 013 NA 010

Guidance 026 047

069 123

mdash 010

Earnings announcement 047 048 044 044 049 051

8-K 173 213

158 192

182 223

Panel C Size of Form 8-K disclosure during one month prior to SEO

8-K size 254 597

297 830

229 481

8-K abnormal size 221 517

267 754 194 399

Panel D Distance between last disclosure and SEO issue date for prior one-month

Annual MEF mdash mdash 1473 1188 mdash mdash Quarterly MEF mdash mdash 1494 904

mdash mdash

All MEF mdash mdash 1456 1052

mdash mdash FWP NA 315 NA 278 NA 345

Guidance mdash mdash 1456 971

mdash mdash Earnings announcement 1440 1346 1329 1439 1500 1307

8-K 813 684 860 837 783 605

Closest disclosure 785 615

824 672 760 586

This table presents mean values for multiple forms of disclosure before and after SOR The forecasting (non-forecasting)

sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The

full sample contains both the forecasting and non-forecasting samples Panel D is for the subgroup of SEO firms with the

respective type of disclosure in the prior one-month period and indicate significance at the 1 5 and 10

level respectively using a standard two-tailed t-test NA = not applicable All variables are defined in the Appendix

41

Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)

Panel A Occurrence frequency and 8-K size during one-month period prior to SEO issue date

Variable

Occurrence of

annual MEF

Occurrence of

guidance

Frequency of

guidance

Frequency of

8-Ks

Frequency of

total disclosure

File size of

last 8-K

File size of

last 8-K

Constant

(1) -189

(2) -125

(3) -002

(4) 085

(5) 154

(6) -1216

(7) -1223

(005) (018) (098) (014) (005) (021) (020)

SOR 043

047

048

038

055

198

492

(009) (006) (004) (001) (001) (006) (004)

Ln(market capitalization) 010 -002 -002 014

010 151 173

(032) (083) (087) (009) (034) (013) (011)

Institutional ownership 110

113

107

-019 036 235 189

(004) (001) (001) (045) (030) (031) (039)

Residual analyst following -002 -006

-004

004

002 001 002

(046) (004) (005) (010) (046) (095) (091)

Analyst dispersion -474

076 038 -011 -014 515 497

(006) (019) (029) (051) (060) (034) (035)

Litigation risk -020 025 -002 -017 -010 -114 -106

(051) (041) (091) (032) (067) (036) (042)

Book-to-market 018 071

024 -015 -001 327 363

(047) (003) (010) (031) (098) (030) (027)

Loss -036 -068

-041

008 -027 -222

-220

(034) (007) (008) (062) (019) (004) (005)

Firm risk 156 391 041 159 340 3324 6409

(086) (064) (094) (065) (049) (024) (011)

Non-SEO 8-K size -690

(014)

Sample Forecasting Forecasting Forecasting Full Full Full Full

Regression Probit Probit OLS OLS OLS OLS OLS

Observations 126 126 126 360 360 358 358

Pseudo R2

0150 0120

R2

010 006 004 006 008

Adjusted R2

003 003 002 004 005

42

Table 4 (continued)

Panel B Distance between last disclosure and SEO issue date by disclosure type

Variable

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Constant

(1) 2456

(2) 1595

(3) 1647

(4) 10149

(5) 1684

(6) 1690

(000) (000) (000) (019) (008) (001)

SOR -563

-083 -106 -1693 -691

-372

(003) (033) (020) (027) (002) (004)

Ln(market capitalization) -112 -104

-102

509 017 -043

(018) (000) (000) (059) (088) (060)

Institutional ownership 350 082 002 -5322

206 -017

(046) (058) (099) (008) (059) (095)

Residual analyst following 001 -023

-018

299 -046

-029

(096) (001) (003) (021) (002) (006)

Analyst dispersion -162 -053 -035 -3270 -157 -156

(064) (060) (070) (018) (053) (041)

Litigation risk -087 020 027 -668 152 216

(070) (083) (076) (073) (062) (026)

Book-to-market 023 034 -031 -633 -372 042

(085) (069) (065) (048) (045) (073)

Loss -020 -096 -085 5399

164 226

(096) (035) (037) (005) (049) (025)

Firm risk -11517

-1358 -2299 -69596 -4750 -6360

(006) (052) (023) (011) (024) (004)

Restricted to one month Yes Yes Yes No No No

Regression OLS OLS OLS OLS OLS OLS

Observations 81 322 332 146 358 360

R2

018 006 006 010 006 005

Adjusted R2

008 003 004 004 003 003

In Panel A the forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The full sample contains both the forecasting

and non-forecasting samples In Panel B Columns 1ndash3 (4ndash6) contain SEOs in which the firm provides a specified disclosure within one month (one year) before

the SEO issue date p-values in parentheses are based on robust standard errors clustered at the firm level and indicate significance at the 1 5 and

10 level respectively from a two-tailed test All variables are defined in the Appendix

43

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 1 Sample distribution and offering characteristics

Panel A Distribution of SEOs by calendar year

Full

Pre- Post- Sample

Sample 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Forecasting 16 15 15 12 17 15 12 24 55 71 126

Non-forecasting 14 23 30 26 19 31 36 55 90 144 234

Full Sample 30 38 45 38 36 46 48 79 145 215 360

Panel B Offering characteristics for SEOs by calendar year

Full

Pre- Post- Sample

Variable 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Shelf (percent) 500 763 822 895 861 826 917 861 752 870 822

Registration days 621 1878 2136 2661 2455 1647 3926 3618 1960 2984 2571

Pre-issue days 34 37 35 26 18 38 20 11 33 20 26

SEO proceeds (millions of dollars) 2372 2232 3564 3400 2805 3508 6239 3235 2872 3937 3508

SEO size (percent) 106 137 127 123 114 140 98 131 123 124 123

Panel C Average number of 8-Ks filed during SEO fiscal year

Full

Pre- Post- Sample

Sample 2002 2003 2004 2005 2006 2007 2008 2009 SOR SOR Period

Forecasting 64 128 133 194 159 141 166 170 128 160 144 Non-forecasting 99 135 143 223 193 189 190 190 155 190 177

Full Sample 81 130 144 216 177 170 185 183 146 179 165

This table presents the distribution values for our full sample of 360 seasoned equity offerings (SEOs) between 2002 and 2009 The Securities

Offering Reform (SOR) became effective December 1 2005 Panels A and B are presented at the SEO level and Panel C is presented at the issuer-

year level We designate SEOs issued between January 1 2002 and November 30 2005 as pre-SOR and SEOs issued between December 1 2005

and December 31 2009 as post-SOR Three SEOs in the 2005 forecasting sample and three SEOs in the 2005 non-forecasting sample occur in

December 2005 These SEOs are included in the 2005 and post-SOR columns The final column presents the total number in Panel A and the

average values in Panels B and C for the full sample period The forecasting (non-forecasting) sample contains SEOs in which the firm has chosen

(not) to provide at least one annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue

date All variables are defined in the Appendix

39

Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Market capitalization (billions of dollars)

Institutional ownership (percent)

522506

6180

601371

7262

888659

6709

850331

7500

298746

5856

478620

7144

Analyst following

Analyst dispersion (percent)

Litigation risk (percent)

Book-to-market

1039

1408

3172

038

1087

2218

2233

044

935

1446

2545

041

1011

797

2254

044

1102

1384

3556

036

1122

2919

2222

044

Loss (percent)

Firm risk (percent)

2897

252

3535

376

1636

273

1690

323

3667

280

4444

402

This table presents the mean values of descriptive statistics for our control variables The forecasting (non-forecasting) sample contains seasoned

equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or quarterly point or range management earnings forecast

(MEF) or both within one year before the SEO issue date The full sample contains both the forecasting and non-forecasting samples and

indicate that the difference in means is significant at the 1 5 and 10 level respectively using a standard two-tailed t-test All variables are

defined in the Appendix

40

Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N= 215) (N = 55) (N = 71) (N = 90) (N = 144)

Panel A Occurrence of guidance disclosure during one month prior to SEO

Annual MEF 1034 1488 2727 4507

mdash mdash Quarterly MEF 1241 1163 3273 3531 mdash mdash All MEF 1724 1953 4545 5915 mdash mdash FWP NA 930 NA 1268 NA 764

Guidance 1724 2605 4545 6338 mdash 764

Panel B Frequency of disclosure during one month prior to SEO

Total disclosure 246 308

271 358

231 283

Annual MEF 012 020 031 059

mdash mdash Quarterly MEF 014 017 038 051 mdash mdash All MEF 026 036 069 110

mdash mdash

FWP NA 011 NA 013 NA 010

Guidance 026 047

069 123

mdash 010

Earnings announcement 047 048 044 044 049 051

8-K 173 213

158 192

182 223

Panel C Size of Form 8-K disclosure during one month prior to SEO

8-K size 254 597

297 830

229 481

8-K abnormal size 221 517

267 754 194 399

Panel D Distance between last disclosure and SEO issue date for prior one-month

Annual MEF mdash mdash 1473 1188 mdash mdash Quarterly MEF mdash mdash 1494 904

mdash mdash

All MEF mdash mdash 1456 1052

mdash mdash FWP NA 315 NA 278 NA 345

Guidance mdash mdash 1456 971

mdash mdash Earnings announcement 1440 1346 1329 1439 1500 1307

8-K 813 684 860 837 783 605

Closest disclosure 785 615

824 672 760 586

This table presents mean values for multiple forms of disclosure before and after SOR The forecasting (non-forecasting)

sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The

full sample contains both the forecasting and non-forecasting samples Panel D is for the subgroup of SEO firms with the

respective type of disclosure in the prior one-month period and indicate significance at the 1 5 and 10

level respectively using a standard two-tailed t-test NA = not applicable All variables are defined in the Appendix

41

Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)

Panel A Occurrence frequency and 8-K size during one-month period prior to SEO issue date

Variable

Occurrence of

annual MEF

Occurrence of

guidance

Frequency of

guidance

Frequency of

8-Ks

Frequency of

total disclosure

File size of

last 8-K

File size of

last 8-K

Constant

(1) -189

(2) -125

(3) -002

(4) 085

(5) 154

(6) -1216

(7) -1223

(005) (018) (098) (014) (005) (021) (020)

SOR 043

047

048

038

055

198

492

(009) (006) (004) (001) (001) (006) (004)

Ln(market capitalization) 010 -002 -002 014

010 151 173

(032) (083) (087) (009) (034) (013) (011)

Institutional ownership 110

113

107

-019 036 235 189

(004) (001) (001) (045) (030) (031) (039)

Residual analyst following -002 -006

-004

004

002 001 002

(046) (004) (005) (010) (046) (095) (091)

Analyst dispersion -474

076 038 -011 -014 515 497

(006) (019) (029) (051) (060) (034) (035)

Litigation risk -020 025 -002 -017 -010 -114 -106

(051) (041) (091) (032) (067) (036) (042)

Book-to-market 018 071

024 -015 -001 327 363

(047) (003) (010) (031) (098) (030) (027)

Loss -036 -068

-041

008 -027 -222

-220

(034) (007) (008) (062) (019) (004) (005)

Firm risk 156 391 041 159 340 3324 6409

(086) (064) (094) (065) (049) (024) (011)

Non-SEO 8-K size -690

(014)

Sample Forecasting Forecasting Forecasting Full Full Full Full

Regression Probit Probit OLS OLS OLS OLS OLS

Observations 126 126 126 360 360 358 358

Pseudo R2

0150 0120

R2

010 006 004 006 008

Adjusted R2

003 003 002 004 005

42

Table 4 (continued)

Panel B Distance between last disclosure and SEO issue date by disclosure type

Variable

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Constant

(1) 2456

(2) 1595

(3) 1647

(4) 10149

(5) 1684

(6) 1690

(000) (000) (000) (019) (008) (001)

SOR -563

-083 -106 -1693 -691

-372

(003) (033) (020) (027) (002) (004)

Ln(market capitalization) -112 -104

-102

509 017 -043

(018) (000) (000) (059) (088) (060)

Institutional ownership 350 082 002 -5322

206 -017

(046) (058) (099) (008) (059) (095)

Residual analyst following 001 -023

-018

299 -046

-029

(096) (001) (003) (021) (002) (006)

Analyst dispersion -162 -053 -035 -3270 -157 -156

(064) (060) (070) (018) (053) (041)

Litigation risk -087 020 027 -668 152 216

(070) (083) (076) (073) (062) (026)

Book-to-market 023 034 -031 -633 -372 042

(085) (069) (065) (048) (045) (073)

Loss -020 -096 -085 5399

164 226

(096) (035) (037) (005) (049) (025)

Firm risk -11517

-1358 -2299 -69596 -4750 -6360

(006) (052) (023) (011) (024) (004)

Restricted to one month Yes Yes Yes No No No

Regression OLS OLS OLS OLS OLS OLS

Observations 81 322 332 146 358 360

R2

018 006 006 010 006 005

Adjusted R2

008 003 004 004 003 003

In Panel A the forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The full sample contains both the forecasting

and non-forecasting samples In Panel B Columns 1ndash3 (4ndash6) contain SEOs in which the firm provides a specified disclosure within one month (one year) before

the SEO issue date p-values in parentheses are based on robust standard errors clustered at the firm level and indicate significance at the 1 5 and

10 level respectively from a two-tailed test All variables are defined in the Appendix

43

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Market capitalization (billions of dollars)

Institutional ownership (percent)

522506

6180

601371

7262

888659

6709

850331

7500

298746

5856

478620

7144

Analyst following

Analyst dispersion (percent)

Litigation risk (percent)

Book-to-market

1039

1408

3172

038

1087

2218

2233

044

935

1446

2545

041

1011

797

2254

044

1102

1384

3556

036

1122

2919

2222

044

Loss (percent)

Firm risk (percent)

2897

252

3535

376

1636

273

1690

323

3667

280

4444

402

This table presents the mean values of descriptive statistics for our control variables The forecasting (non-forecasting) sample contains seasoned

equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or quarterly point or range management earnings forecast

(MEF) or both within one year before the SEO issue date The full sample contains both the forecasting and non-forecasting samples and

indicate that the difference in means is significant at the 1 5 and 10 level respectively using a standard two-tailed t-test All variables are

defined in the Appendix

40

Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N= 215) (N = 55) (N = 71) (N = 90) (N = 144)

Panel A Occurrence of guidance disclosure during one month prior to SEO

Annual MEF 1034 1488 2727 4507

mdash mdash Quarterly MEF 1241 1163 3273 3531 mdash mdash All MEF 1724 1953 4545 5915 mdash mdash FWP NA 930 NA 1268 NA 764

Guidance 1724 2605 4545 6338 mdash 764

Panel B Frequency of disclosure during one month prior to SEO

Total disclosure 246 308

271 358

231 283

Annual MEF 012 020 031 059

mdash mdash Quarterly MEF 014 017 038 051 mdash mdash All MEF 026 036 069 110

mdash mdash

FWP NA 011 NA 013 NA 010

Guidance 026 047

069 123

mdash 010

Earnings announcement 047 048 044 044 049 051

8-K 173 213

158 192

182 223

Panel C Size of Form 8-K disclosure during one month prior to SEO

8-K size 254 597

297 830

229 481

8-K abnormal size 221 517

267 754 194 399

Panel D Distance between last disclosure and SEO issue date for prior one-month

Annual MEF mdash mdash 1473 1188 mdash mdash Quarterly MEF mdash mdash 1494 904

mdash mdash

All MEF mdash mdash 1456 1052

mdash mdash FWP NA 315 NA 278 NA 345

Guidance mdash mdash 1456 971

mdash mdash Earnings announcement 1440 1346 1329 1439 1500 1307

8-K 813 684 860 837 783 605

Closest disclosure 785 615

824 672 760 586

This table presents mean values for multiple forms of disclosure before and after SOR The forecasting (non-forecasting)

sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The

full sample contains both the forecasting and non-forecasting samples Panel D is for the subgroup of SEO firms with the

respective type of disclosure in the prior one-month period and indicate significance at the 1 5 and 10

level respectively using a standard two-tailed t-test NA = not applicable All variables are defined in the Appendix

41

Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)

Panel A Occurrence frequency and 8-K size during one-month period prior to SEO issue date

Variable

Occurrence of

annual MEF

Occurrence of

guidance

Frequency of

guidance

Frequency of

8-Ks

Frequency of

total disclosure

File size of

last 8-K

File size of

last 8-K

Constant

(1) -189

(2) -125

(3) -002

(4) 085

(5) 154

(6) -1216

(7) -1223

(005) (018) (098) (014) (005) (021) (020)

SOR 043

047

048

038

055

198

492

(009) (006) (004) (001) (001) (006) (004)

Ln(market capitalization) 010 -002 -002 014

010 151 173

(032) (083) (087) (009) (034) (013) (011)

Institutional ownership 110

113

107

-019 036 235 189

(004) (001) (001) (045) (030) (031) (039)

Residual analyst following -002 -006

-004

004

002 001 002

(046) (004) (005) (010) (046) (095) (091)

Analyst dispersion -474

076 038 -011 -014 515 497

(006) (019) (029) (051) (060) (034) (035)

Litigation risk -020 025 -002 -017 -010 -114 -106

(051) (041) (091) (032) (067) (036) (042)

Book-to-market 018 071

024 -015 -001 327 363

(047) (003) (010) (031) (098) (030) (027)

Loss -036 -068

-041

008 -027 -222

-220

(034) (007) (008) (062) (019) (004) (005)

Firm risk 156 391 041 159 340 3324 6409

(086) (064) (094) (065) (049) (024) (011)

Non-SEO 8-K size -690

(014)

Sample Forecasting Forecasting Forecasting Full Full Full Full

Regression Probit Probit OLS OLS OLS OLS OLS

Observations 126 126 126 360 360 358 358

Pseudo R2

0150 0120

R2

010 006 004 006 008

Adjusted R2

003 003 002 004 005

42

Table 4 (continued)

Panel B Distance between last disclosure and SEO issue date by disclosure type

Variable

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Constant

(1) 2456

(2) 1595

(3) 1647

(4) 10149

(5) 1684

(6) 1690

(000) (000) (000) (019) (008) (001)

SOR -563

-083 -106 -1693 -691

-372

(003) (033) (020) (027) (002) (004)

Ln(market capitalization) -112 -104

-102

509 017 -043

(018) (000) (000) (059) (088) (060)

Institutional ownership 350 082 002 -5322

206 -017

(046) (058) (099) (008) (059) (095)

Residual analyst following 001 -023

-018

299 -046

-029

(096) (001) (003) (021) (002) (006)

Analyst dispersion -162 -053 -035 -3270 -157 -156

(064) (060) (070) (018) (053) (041)

Litigation risk -087 020 027 -668 152 216

(070) (083) (076) (073) (062) (026)

Book-to-market 023 034 -031 -633 -372 042

(085) (069) (065) (048) (045) (073)

Loss -020 -096 -085 5399

164 226

(096) (035) (037) (005) (049) (025)

Firm risk -11517

-1358 -2299 -69596 -4750 -6360

(006) (052) (023) (011) (024) (004)

Restricted to one month Yes Yes Yes No No No

Regression OLS OLS OLS OLS OLS OLS

Observations 81 322 332 146 358 360

R2

018 006 006 010 006 005

Adjusted R2

008 003 004 004 003 003

In Panel A the forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The full sample contains both the forecasting

and non-forecasting samples In Panel B Columns 1ndash3 (4ndash6) contain SEOs in which the firm provides a specified disclosure within one month (one year) before

the SEO issue date p-values in parentheses are based on robust standard errors clustered at the firm level and indicate significance at the 1 5 and

10 level respectively from a two-tailed test All variables are defined in the Appendix

43

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N= 215) (N = 55) (N = 71) (N = 90) (N = 144)

Panel A Occurrence of guidance disclosure during one month prior to SEO

Annual MEF 1034 1488 2727 4507

mdash mdash Quarterly MEF 1241 1163 3273 3531 mdash mdash All MEF 1724 1953 4545 5915 mdash mdash FWP NA 930 NA 1268 NA 764

Guidance 1724 2605 4545 6338 mdash 764

Panel B Frequency of disclosure during one month prior to SEO

Total disclosure 246 308

271 358

231 283

Annual MEF 012 020 031 059

mdash mdash Quarterly MEF 014 017 038 051 mdash mdash All MEF 026 036 069 110

mdash mdash

FWP NA 011 NA 013 NA 010

Guidance 026 047

069 123

mdash 010

Earnings announcement 047 048 044 044 049 051

8-K 173 213

158 192

182 223

Panel C Size of Form 8-K disclosure during one month prior to SEO

8-K size 254 597

297 830

229 481

8-K abnormal size 221 517

267 754 194 399

Panel D Distance between last disclosure and SEO issue date for prior one-month

Annual MEF mdash mdash 1473 1188 mdash mdash Quarterly MEF mdash mdash 1494 904

mdash mdash

All MEF mdash mdash 1456 1052

mdash mdash FWP NA 315 NA 278 NA 345

Guidance mdash mdash 1456 971

mdash mdash Earnings announcement 1440 1346 1329 1439 1500 1307

8-K 813 684 860 837 783 605

Closest disclosure 785 615

824 672 760 586

This table presents mean values for multiple forms of disclosure before and after SOR The forecasting (non-forecasting)

sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The

full sample contains both the forecasting and non-forecasting samples Panel D is for the subgroup of SEO firms with the

respective type of disclosure in the prior one-month period and indicate significance at the 1 5 and 10

level respectively using a standard two-tailed t-test NA = not applicable All variables are defined in the Appendix

41

Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)

Panel A Occurrence frequency and 8-K size during one-month period prior to SEO issue date

Variable

Occurrence of

annual MEF

Occurrence of

guidance

Frequency of

guidance

Frequency of

8-Ks

Frequency of

total disclosure

File size of

last 8-K

File size of

last 8-K

Constant

(1) -189

(2) -125

(3) -002

(4) 085

(5) 154

(6) -1216

(7) -1223

(005) (018) (098) (014) (005) (021) (020)

SOR 043

047

048

038

055

198

492

(009) (006) (004) (001) (001) (006) (004)

Ln(market capitalization) 010 -002 -002 014

010 151 173

(032) (083) (087) (009) (034) (013) (011)

Institutional ownership 110

113

107

-019 036 235 189

(004) (001) (001) (045) (030) (031) (039)

Residual analyst following -002 -006

-004

004

002 001 002

(046) (004) (005) (010) (046) (095) (091)

Analyst dispersion -474

076 038 -011 -014 515 497

(006) (019) (029) (051) (060) (034) (035)

Litigation risk -020 025 -002 -017 -010 -114 -106

(051) (041) (091) (032) (067) (036) (042)

Book-to-market 018 071

024 -015 -001 327 363

(047) (003) (010) (031) (098) (030) (027)

Loss -036 -068

-041

008 -027 -222

-220

(034) (007) (008) (062) (019) (004) (005)

Firm risk 156 391 041 159 340 3324 6409

(086) (064) (094) (065) (049) (024) (011)

Non-SEO 8-K size -690

(014)

Sample Forecasting Forecasting Forecasting Full Full Full Full

Regression Probit Probit OLS OLS OLS OLS OLS

Observations 126 126 126 360 360 358 358

Pseudo R2

0150 0120

R2

010 006 004 006 008

Adjusted R2

003 003 002 004 005

42

Table 4 (continued)

Panel B Distance between last disclosure and SEO issue date by disclosure type

Variable

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Constant

(1) 2456

(2) 1595

(3) 1647

(4) 10149

(5) 1684

(6) 1690

(000) (000) (000) (019) (008) (001)

SOR -563

-083 -106 -1693 -691

-372

(003) (033) (020) (027) (002) (004)

Ln(market capitalization) -112 -104

-102

509 017 -043

(018) (000) (000) (059) (088) (060)

Institutional ownership 350 082 002 -5322

206 -017

(046) (058) (099) (008) (059) (095)

Residual analyst following 001 -023

-018

299 -046

-029

(096) (001) (003) (021) (002) (006)

Analyst dispersion -162 -053 -035 -3270 -157 -156

(064) (060) (070) (018) (053) (041)

Litigation risk -087 020 027 -668 152 216

(070) (083) (076) (073) (062) (026)

Book-to-market 023 034 -031 -633 -372 042

(085) (069) (065) (048) (045) (073)

Loss -020 -096 -085 5399

164 226

(096) (035) (037) (005) (049) (025)

Firm risk -11517

-1358 -2299 -69596 -4750 -6360

(006) (052) (023) (011) (024) (004)

Restricted to one month Yes Yes Yes No No No

Regression OLS OLS OLS OLS OLS OLS

Observations 81 322 332 146 358 360

R2

018 006 006 010 006 005

Adjusted R2

008 003 004 004 003 003

In Panel A the forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The full sample contains both the forecasting

and non-forecasting samples In Panel B Columns 1ndash3 (4ndash6) contain SEOs in which the firm provides a specified disclosure within one month (one year) before

the SEO issue date p-values in parentheses are based on robust standard errors clustered at the firm level and indicate significance at the 1 5 and

10 level respectively from a two-tailed test All variables are defined in the Appendix

43

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)

Panel A Occurrence frequency and 8-K size during one-month period prior to SEO issue date

Variable

Occurrence of

annual MEF

Occurrence of

guidance

Frequency of

guidance

Frequency of

8-Ks

Frequency of

total disclosure

File size of

last 8-K

File size of

last 8-K

Constant

(1) -189

(2) -125

(3) -002

(4) 085

(5) 154

(6) -1216

(7) -1223

(005) (018) (098) (014) (005) (021) (020)

SOR 043

047

048

038

055

198

492

(009) (006) (004) (001) (001) (006) (004)

Ln(market capitalization) 010 -002 -002 014

010 151 173

(032) (083) (087) (009) (034) (013) (011)

Institutional ownership 110

113

107

-019 036 235 189

(004) (001) (001) (045) (030) (031) (039)

Residual analyst following -002 -006

-004

004

002 001 002

(046) (004) (005) (010) (046) (095) (091)

Analyst dispersion -474

076 038 -011 -014 515 497

(006) (019) (029) (051) (060) (034) (035)

Litigation risk -020 025 -002 -017 -010 -114 -106

(051) (041) (091) (032) (067) (036) (042)

Book-to-market 018 071

024 -015 -001 327 363

(047) (003) (010) (031) (098) (030) (027)

Loss -036 -068

-041

008 -027 -222

-220

(034) (007) (008) (062) (019) (004) (005)

Firm risk 156 391 041 159 340 3324 6409

(086) (064) (094) (065) (049) (024) (011)

Non-SEO 8-K size -690

(014)

Sample Forecasting Forecasting Forecasting Full Full Full Full

Regression Probit Probit OLS OLS OLS OLS OLS

Observations 126 126 126 360 360 358 358

Pseudo R2

0150 0120

R2

010 006 004 006 008

Adjusted R2

003 003 002 004 005

42

Table 4 (continued)

Panel B Distance between last disclosure and SEO issue date by disclosure type

Variable

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Constant

(1) 2456

(2) 1595

(3) 1647

(4) 10149

(5) 1684

(6) 1690

(000) (000) (000) (019) (008) (001)

SOR -563

-083 -106 -1693 -691

-372

(003) (033) (020) (027) (002) (004)

Ln(market capitalization) -112 -104

-102

509 017 -043

(018) (000) (000) (059) (088) (060)

Institutional ownership 350 082 002 -5322

206 -017

(046) (058) (099) (008) (059) (095)

Residual analyst following 001 -023

-018

299 -046

-029

(096) (001) (003) (021) (002) (006)

Analyst dispersion -162 -053 -035 -3270 -157 -156

(064) (060) (070) (018) (053) (041)

Litigation risk -087 020 027 -668 152 216

(070) (083) (076) (073) (062) (026)

Book-to-market 023 034 -031 -633 -372 042

(085) (069) (065) (048) (045) (073)

Loss -020 -096 -085 5399

164 226

(096) (035) (037) (005) (049) (025)

Firm risk -11517

-1358 -2299 -69596 -4750 -6360

(006) (052) (023) (011) (024) (004)

Restricted to one month Yes Yes Yes No No No

Regression OLS OLS OLS OLS OLS OLS

Observations 81 322 332 146 358 360

R2

018 006 006 010 006 005

Adjusted R2

008 003 004 004 003 003

In Panel A the forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The full sample contains both the forecasting

and non-forecasting samples In Panel B Columns 1ndash3 (4ndash6) contain SEOs in which the firm provides a specified disclosure within one month (one year) before

the SEO issue date p-values in parentheses are based on robust standard errors clustered at the firm level and indicate significance at the 1 5 and

10 level respectively from a two-tailed test All variables are defined in the Appendix

43

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 4 (continued)

Panel B Distance between last disclosure and SEO issue date by disclosure type

Variable

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Distance of

last guidance

Distance of

last 8-K

Distance of

closest disclosure

Constant

(1) 2456

(2) 1595

(3) 1647

(4) 10149

(5) 1684

(6) 1690

(000) (000) (000) (019) (008) (001)

SOR -563

-083 -106 -1693 -691

-372

(003) (033) (020) (027) (002) (004)

Ln(market capitalization) -112 -104

-102

509 017 -043

(018) (000) (000) (059) (088) (060)

Institutional ownership 350 082 002 -5322

206 -017

(046) (058) (099) (008) (059) (095)

Residual analyst following 001 -023

-018

299 -046

-029

(096) (001) (003) (021) (002) (006)

Analyst dispersion -162 -053 -035 -3270 -157 -156

(064) (060) (070) (018) (053) (041)

Litigation risk -087 020 027 -668 152 216

(070) (083) (076) (073) (062) (026)

Book-to-market 023 034 -031 -633 -372 042

(085) (069) (065) (048) (045) (073)

Loss -020 -096 -085 5399

164 226

(096) (035) (037) (005) (049) (025)

Firm risk -11517

-1358 -2299 -69596 -4750 -6360

(006) (052) (023) (011) (024) (004)

Restricted to one month Yes Yes Yes No No No

Regression OLS OLS OLS OLS OLS OLS

Observations 81 322 332 146 358 360

R2

018 006 006 010 006 005

Adjusted R2

008 003 004 004 003 003

In Panel A the forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one annual or

quarterly point or range management earnings forecast (MEF) or both within one year before the SEO issue date The full sample contains both the forecasting

and non-forecasting samples In Panel B Columns 1ndash3 (4ndash6) contain SEOs in which the firm provides a specified disclosure within one month (one year) before

the SEO issue date p-values in parentheses are based on robust standard errors clustered at the firm level and indicate significance at the 1 5 and

10 level respectively from a two-tailed test All variables are defined in the Appendix

43

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 5 Univariate comparison of error and bias in management earnings forecasts

(MEFs) before and after the Securities Offering Reform (SOR)

Panel A Last MEF during one month prior to SEO

Pre-SOR Post-SOR

Variable (N = 25) (N = 42) p-value

Error 243 072 000

Bias 133 023 009

Panel B First annual MEF during one year versus last annual MEF during one month prior to

SEO

Pre-SOR Post-SOR

Variable (N = 18) (N = 35) p-value

Error

First annual MEF 266 136 008

Last annual MEF 247 077 001

Difference -020 -059

039

Bias

First annual MEF 182 071 020

Last annual MEF 178 032 005

Difference -003 -040

046

Horizon

First annual MEF 33139 33837 085

Last annual MEF 21567 17494 020

Panel C First quarterly MEF during one year versus last quarterly MEF during one month prior

to SEO

Pre-SOR Post-SOR

Variable (N = 21) (N = 29) p-value

Error

First quarterly MEF 136 093 030

Last quarterly MEF 234 084 003

Difference 098 -009 006

Bias

First quarterly MEF 051 -027 014

Last quarterly MEF 106 -039 007

Difference 055 -012 031

Horizon

First quarterly MEF 4319 2917 017

Last quarterly MEF 3433 3066 069

Panel A presents the mean values of error and bias for seasoned equity offerings (SEOs) in which the

firm provides an MEF within one month before the SEO issue date Error and bias are annualized for

quarterly MEFs Values for annual MEFs are used if both annual and quarterly MEFs are provided

Panels B and C present the mean values of the first annual or quarterly MEF provided within one year

before the SEO issue date and the last annual or quarterly MEF provided within one month before the

SEO issue date In Columns 1ndash2 and indicate if difference in first and last error or bias is

significantly different from zero at the 1 5 and 10 level respectively using a standard two-tailed t-

test In Column 3 and indicate if difference in pre- and post-SOR means is significant at the 1

5 and 10 level respectively using a standard two-tailed t-test Variables are defined in the Appendix

44

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs)

before and after the Securities Offering Reform (SOR)

MEF error MEF error MEF bias MEF bias

Variable (1) (2) (3) (4)

Constant -015 061 -303 -200

(093) (067) (014) (020)

SOR -156

-104

-149

-136

(003) (003) (006) (001)

Ln(market capitalization) 025 001 043

027

(026) (097) (008) (016)

Institutional ownership -001 123 137 136

(099) (012) (033) (017)

Residual analyst following 009

-003 009 005

(009) (053) (015) (043)

Analyst dispersion -027 091 -231

-189

(081) (043) (005) (004)

Litigation risk -066 -087

-108

-103

(011) (001) (003) (001)

Book-to-market 156

054 189

072

(010) (021) (007) (012)

Loss -002 092 -038 114

(098) (025) (075) (020)

Firm risk -240 -934 293 177

(092) (041) (091) (089)

Horizon 046 072

-018 034

(053) (002) (084) (051)

Sample Forecasting Forecasting Forecasting Forecasting

Restricted to one month Yes No Yes No

Regression OLS OLS OLS OLS

Observations 67 126 67 126

R2

025 018 024 015

Adjusted R2

012 010 010 007

The forecasting sample in Columns 1 and 3 (2 and 4) contains seasoned equity offerings (SEOs) in which

the firm has chosen to provide at least one annual or quarterly point or range MEF or both within one

month (one year) before the SEO issue date Error and bias are based on the last annual or annualized

quarterly MEF provided during the specified time period p-values in parentheses are based on robust

standard errors clustered at the firm level and indicate significance at the 1 5 and 10 level

respectively from a two-tailed test Variables are defined in the Appendix

45

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering

Reform (SOR)

Panel A Pre- and post-SOR magnitude

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Magnitude [-200] 3756 5520 3650 4332 3822 6106

(000) (006) (000)

Magnitude [1+20] 2906 3918 2755 3117 2991 4313

(000) (018) (000)

Abnormal magnitude 855 1602 895 1216 830 1793

(000) (022) (000)

Panel B Pre-SOR magnitude and post-SOR magnitude by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Magnitude [-200]

(N = 145)

3756

(N = 109)

5831

(N = 106)

5201

(N = 101)

5936

(N = 114)

5151

Magnitude [1+20] 2906

(000)

3823

(000)

4016

(000)

3835

(000)

3992

Abnormal magnitude 855

(000)

2008

(000)

1185

(000)

2102

(000)

1160

(000) (022) (000) (024)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

management earnings forecast (MEF) within one year before the SEO issue date Magnitude is the cumulative absolute value of daily return net of

the CRSP value-weighted index return over the specified trading days with day 0 being the volume-based corrected SEO issue date (Corwin

2003) Abnormal magnitude is magnitude [-20 0] less magnitude [+1+20] Panel A tests for statistical differences between mean values of

magnitude for the stated pre- and post-SOR samples Panel B tests for statistical differences between mean values of magnitude for the stated post-

SOR subsample and the pre-SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an

8-K) within one week prior to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a

standard two-tailed t-test p-values from these tests are displayed in parentheses Variables are defined in the Appendix

46

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Table 8 Stock returns before and after the Securities Offering Reform (SOR)

Panel A Pre- and post-SOR return

Full sample Forecasting Non-forecasting

Pre-SOR Post-SOR Pre-SOR Post-SOR Pre-SOR Post-SOR

Variable (N = 145) (N = 215) (N = 55) (N = 71) (N = 90) (N = 144)

Return [-200] 245 503 200 072 272 716

(025) (057) (017)

Return [1+20] 160 031 118 -071 185 082

(030) (031) (053)

Abnormal return 085 472 082 143 087 634

(011) (082) (010)

Panel B Pre-SOR return and post-SOR return by disclosure within one week of the SEO issue date

Pre-SOR Post-SOR Post-SOR

Full sample Short distance disclosure Short distance 8-K

Yes No Yes No

Variable

Return [-200]

(N = 145)

245

(N = 109)

836

(N = 106)

161

(N = 101)

936

(N = 114)

120

(009) (072) (007) (059)

Return [1+20] 160 088 -027 143 -038

Abnormal return 085

(065)

748

(022)

188

(092)

793

(014)

188

(006) (072) (006) (070)

The forecasting (non-forecasting) sample contains seasoned equity offerings (SEOs) in which the firm has chosen (not) to provide at least one

annual or quarterly point or range management earnings forecast (MEF) or both within one year before the SEO iss ue date Return is the

cumulative raw daily returns over the specified number of trading days with day 0 being the volume -based corrected SEO issue date (Corwin

2003) Abnormal return is return [-200] less return [+1+20] Panel A tests for statistical differences between mean values of return for the stated

pre- and post-SOR samples Panel B tests for statistical differences between mean values of return for the stated post-SOR subsample and the pre-

SOR full sample Short distance disclosure (8-K) contains SEOs in which the firm provides any form of disclosure (an 8-K) within one week prior

to the SEO issue date and indicate significant differences at the 1 5 and 10 level respectively using a standard two-tailed t-test p-

values from these tests are displayed in parentheses Variables are defined in the Appendix

47

• Cover page
• • DERA Working Paper Header
  • Information Environment under SOR
  • • Information Environment under SOR Rev3
    • • 1 Introduction
      • 2 Background and empirical questions
      • • 21 Before the Securities Offering Reform of 2005
        • 22 After the Securities Offering Reform of 2005
        • 23 Empirical tests
        • 24 Capital formation period
        • • 3 Data and sample selection
          • • 31 Disclosures analyzed
            • 32 Sample selection
            • • 4 Research design
              • • 41 Occurrence frequency 8-K size and distance
                • 42 Management earnings forecast accuracy
                • 43 Information magnitude and stock returns
                • 44 Comparison with other studies
                • • 5 Results
                  • • 51 Descriptive statistics
                    • 52 Results for occurrence frequency 8-K size and distance
                    • 53 Results for management earnings forecast accuracy
                    • 54 Results for information magnitude and stock returns
                    • • 6 Conclusion
                      • Appendix Variable definitions and other key terms
                      • References
                      • Fig 1 Timeline of shelf registration and issue before and after the Securities Offering Reform (SOR)
                      • Fig 2 Cumulative magnitude over one month prior to seasoned equity offering (SEO)
                      • Fig 3 Aggregate magnitude for all 8-Ks filed during one month prior to seasoned equity offering (SEO)
                      • Table 1 Sample distribution and offering characteristics
                      • Table 2 Descriptive statistics before and after the Securities Offering Reform (SOR)
                      • Table 3 Univariate comparison of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 4 Multivariate analysis of disclosure before and after the Securities Offering Reform (SOR)
                      • Table 5 Univariate comparison of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 6 Multivariate analysis of error and bias in management earnings forecasts (MEFs) before and after the Securities Offering Reform (SOR)
                      • Table 7 Magnitude of information reflected in absolute market-adjusted returns before and after the Securities Offering Reform (SOR)
                      • Table 8 Stock returns before and after the Securities Offering Reform (SOR)